{"schema_version":1,"generated_at":"2026-10-26T10:18:14Z","issues_generated_at":"2026-10-26T10:12:02Z","lookback_days":45,"license_note":"Issue metadata from public GitHub repos.","issues":[{"id":"anthropics/claude-code#46917","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":46917,"title":"CC v2.1.100+ inflates cache_creation by ~20K tokens vs v2.1.98 — same payload, server-side","url":"https://github.com/anthropics/claude-code/issues/46917","state":"open","state_reason":null,"labels":["bug","has repro","platform:linux","area:cost","platform:wsl"],"created":"2026-04-12T08:14:34Z","updated":"2026-10-08T15:25:36Z","closed":null,"reactions":219,"thumbs_up":215,"comments":43,"severity":"high","severity_reasons":["engagement 262"],"topic":"self-host/headless","classes":["self-host/headless","auth/session","network"],"cosmetic_ui":false,"summary":"Claude Code versions 2.1.100 and 2.1.101 consume ~20,000 more cachecreationinputtokens per request than v2.1.98, despite sending fewer bytes in the request payload. The inflation is server-side and version-specific (likely User-Agent…","versions":["2.1.100","2.1.98","2.1.101","1.0.2"],"error_snippets":["export ANTHROPIC_BASE_URL=\"http://localhost:8000\"","This is not a billing-only issue — these tokens enter the model's context window and may affect output quality.","Each test is a **cold cache, single API call, no session state** (`--print` mode). Same machine, same project, same account, minutes apart.","Cross-account test (same v2.1.98, two different Max accounts): delta < 500 tokens = noise. **Not account-specific.**","The 20K extra tokens are `cache_creation_input_tokens` — this means they enter the model's context window, not just the billing ledger. If the server injects additional content invisible to the user:"],"workaround":{"text":"Downgrade to v2.1.98 or earlier. Check `~/.local/share/claude/versions/` for available binaries, or use `npx claude-code@2.1.98`.","source":"issue body","url":"https://github.com/anthropics/claude-code/issues/46917"},"fixed_in":null,"score":482.2,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openai/codex#25719","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":25719,"title":"Codex Desktop for macOS repeatedly triggers `syspolicyd` / `trustd` CPU and memory runaway","url":"https://github.com/openai/codex/issues/25719","state":"open","state_reason":null,"labels":["bug","app","computer-use","performance"],"created":"2026-06-01T23:59:29Z","updated":"2026-09-15T11:24:55Z","closed":null,"reactions":455,"thumbs_up":399,"comments":90,"severity":"high","severity_reasons":["engagement 545"],"topic":"crash/hang","classes":["crash/hang"],"cosmetic_ui":false,"summary":"Codex Desktop: 26.527.60818 Codex build: 3437 Darwin 25.5.0 arm64 arm Codex Desktop for macOS triggers a persistent macOS Gatekeeper/SystemPolicy loop after launch. With Codex closed, the system is quiet. As soon as Codex Desktop is…","versions":["26.527.60818","26.527.31326"],"error_snippets":["Codex Desktop: 26.527.60818","Timezone: Europe/Rome","Codex Desktop: 26.527.31326","11.3 29728 3.0 756064 /Applications/Codex.app/Contents/Resources/codex app-server --analytics-default-enabled","\"Codex Computer Use\" is damaged and can't be opened."],"workaround":null,"fixed_in":null,"score":418.0,"first_seen":"2026-10-18T04:31:06Z"},{"id":"Aider-AI/aider#3314","harness":"aider","harness_name":"Aider","repo":"Aider-AI/aider","number":3314,"title":"MCP SUPPORT","url":"https://github.com/Aider-AI/aider/issues/3314","state":"open","state_reason":null,"labels":[],"created":"2025-02-20T12:03:07Z","updated":"2026-10-22T19:33:32Z","closed":null,"reactions":244,"thumbs_up":196,"comments":25,"severity":"high","severity_reasons":["engagement 269"],"topic":"mcp/tools","classes":["mcp/tools"],"cosmetic_ui":false,"summary":"Is there plane to add mcp server call?","versions":[],"error_snippets":["A. Forcing the user to manually copy paste documentation, either into the context window, or into local project files","or B. Hallucinating things that SOTA models don't even get wrong anymore, because all the SOTA models now have tool calling in the CoT.","> It already exists. https://github.com/lutzleonhardt/mcpm-aider","I wouldn't say it already exists, there is more interest in having aider use MCP servers directly.","It already exists."],"workaround":null,"fixed_in":null,"score":370.6,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openai/codex#28879","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":28879,"title":"Codex (gpt-5.5, Plus plan) — rate-limit cost per token jumped ~10-20x since June 16, draining the 5h budget in 2-3 prompts","url":"https://github.com/openai/codex/issues/28879","state":"closed","state_reason":"completed","labels":["bug","rate-limits","app"],"created":"2026-06-18T08:02:01Z","updated":"2026-10-02T08:55:47Z","closed":"2026-08-23T04:43:07Z","reactions":561,"thumbs_up":365,"comments":211,"severity":"high","severity_reasons":["engagement 772"],"topic":"auth/session","classes":["auth/session","crash/hang"],"cosmetic_ui":false,"summary":"Starting ~June 16, my ChatGPT Plus Codex budget drains in 2–3 prompts on gpt-5.5, where the same model/plan/app previously gave 20+ prompts. My Codex session logs (tokencount / ratelimits events) show the limit-% consumed per token has…","versions":[],"error_snippets":["Starting ~June 16, my ChatGPT Plus Codex budget drains in 2–3 prompts on `gpt-5.5`, where the same model/plan/app previously gave 20+ prompts. My Codex session logs (`token_count` / `rate_limits` events) show the **limit","OpenAI status page shows an active incident ongoing ~2 days (since ~June 16), which matches exactly when my usable prompt count dropped from ~9 sessions/day to ~4."],"workaround":null,"fixed_in":null,"score":369.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"anthropics/claude-code#28240","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":28240,"title":"[BUG] Permission prompt incorrectly triggers on cd instead of the actual command in compound bash statements","url":"https://github.com/anthropics/claude-code/issues/28240","state":"open","state_reason":null,"labels":["bug","platform:windows","area:bash","regression","area:permissions"],"created":"2026-02-24T16:42:57Z","updated":"2026-10-18T12:57:18Z","closed":null,"reactions":209,"thumbs_up":209,"comments":49,"severity":"high","severity_reasons":["label:regression","engagement 258"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"When Claude Code generates a compound bash command using && (e.g.: cd /some/path && git add file && git commit -m \"msg\"), the permission approval prompt incorrectly identifies cd as the action requiring approval, showing a cd: permission…","versions":["2.1.40","2.1.52"],"error_snippets":["The problem repeats on every execution. The cd:* prompt keeps appearing and cannot be whitelisted for the session, making the workflow unusable.","In earlier versions, the same compound command correctly prompted for git add and git commit as the actions requiring permission, and allowed the user to whitelist them for the session.","5. Approve it. The prompt appears again on the next execution, it cannot be whitelisted"],"workaround":null,"fixed_in":null,"score":357.3,"first_seen":"2026-10-18T04:30:35Z"},{"id":"openai/codex#28058","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":28058,"title":"Regression: encrypted MultiAgentV2 messages remove readable task audit trail","url":"https://github.com/openai/codex/issues/28058","state":"open","state_reason":null,"labels":["bug","CLI","subagent"],"created":"2026-06-13T08:06:35Z","updated":"2026-10-13T14:54:17Z","closed":null,"reactions":151,"thumbs_up":137,"comments":40,"severity":"high","severity_reasons":["title:breakage","engagement 191"],"topic":"install/update","classes":["install/update","crash/hang","auth/session"],"cosmetic_ui":false,"summary":"Upstream main after #26210 (Encrypt multi-agent v2 message payloads, merged 2026-06-05). This appears to affect versions that include that change and enable MultiAgentV2 (post-0.137.0). Not subscription-specific. Not model-specific. This…","versions":["0.137.0"],"error_snippets":["message_delivery = \"plaintext\" # encrypted | encrypted_with_audit | plaintext","parameter as encrypted, storing only","as an empty string and stores the payload only in","plus the encrypted","is empty, it records"],"workaround":null,"fixed_in":null,"score":346.2,"first_seen":"2026-10-18T04:31:06Z"},{"id":"openai/codex#48074","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":48074,"title":"Windows: terminal windows repeatedly flash during requests after installing the Codex daemon","url":"https://github.com/openai/codex/issues/48074","state":"closed","state_reason":"completed","labels":["bug","windows-os","CLI","app-server"],"created":"2026-09-25T11:48:04Z","updated":"2026-10-15T10:59:50Z","closed":"2026-10-01T15:30:51Z","reactions":164,"thumbs_up":155,"comments":153,"severity":"high","severity_reasons":["engagement 317"],"topic":"always-on/daemon","classes":["always-on/daemon","self-host/headless","install/update","network"],"cosmetic_ui":false,"summary":"codex-cli 0.157.0 win terminal cmd Description: After installing the Codex daemon, several terminal windows briefly open and close during every request. This happens repeatedly and is distracting. Codex daemon version: 0.157.0 Client…","versions":["0.157.0"],"error_snippets":["After installing the Codex daemon, several terminal windows briefly open and close during every request. This happens repeatedly and is distracting.","Windows sandbox setting: elevated","Expected behavior:","Several terminal windows flash on screen during each request. I started noticing this after installing the daemon."],"workaround":null,"fixed_in":null,"score":344.3,"first_seen":"2026-10-18T04:31:06Z"},{"id":"openai/codex#13041","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":13041,"title":"WebSocket upgrade succeeds then server closes with 1008 Policy (falls back to HTTPS)","url":"https://github.com/openai/codex/issues/13041","state":"closed","state_reason":"completed","labels":["bug","connectivity"],"created":"2026-02-27T18:38:44Z","updated":"2026-09-14T06:29:01Z","closed":"2026-08-05T03:09:41Z","reactions":170,"thumbs_up":170,"comments":75,"severity":"high","severity_reasons":["engagement 245"],"topic":"self-host/headless","classes":["self-host/headless","install/update","network","auth/session"],"cosmetic_ui":false,"summary":"Codex websocket transport to wss://chatgpt.com/backend-api/codex/responses upgrades successfully and is then immediately closed by server policy (1008), leading to reconnect loop and fallback to HTTPS. Codex CLI: 0.106.0 (also reproduced…","versions":["0.106.0","0.107.0-alpha.5"],"error_snippets":["codex exec -C /home/arch --skip-git-repo-check --json \"Reply with exactly: WS_FINAL_CHECK\""],"workaround":{"text":"reverted to the previous version","source":"comment by @daimaidaibuff (1 reactions)","url":"https://github.com/openai/codex/issues/13041#issuecomment-3977214991"},"fixed_in":null,"score":321.9,"first_seen":"2026-10-18T04:31:06Z"},{"id":"openai/codex#17265","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":17265,"title":"Codex does not auto-refresh routed MCP OAuth tokens even when a refresh token is stored","url":"https://github.com/openai/codex/issues/17265","state":"open","state_reason":null,"labels":["bug","auth","mcp"],"created":"2026-04-09T22:27:30Z","updated":"2026-09-12T14:35:28Z","closed":null,"reactions":69,"thumbs_up":69,"comments":38,"severity":"medium","severity_reasons":["engagement 107","class:auth/session","class:mcp/tools"],"topic":"auth/session","classes":["auth/session","mcp/tools","self-host/headless"],"cosmetic_ui":false,"summary":"Codex persists a refreshtoken for a routed MCP server in ~/.codex/.credentials.json, but it does not automatically refresh the access token when the token expires. As a result, once the access token expires: MCP tool calls start failing…","versions":[],"error_snippets":["Codex persists a `refresh_token` for a routed MCP server in `~/.codex/.credentials.json`, but it does not automatically refresh the access token when the token expires.","As a result, once the access token expires:","If Codex has a valid stored `refresh_token` for the MCP server, it should automatically refresh the expired access token and continue making authenticated MCP calls without requiring manual CLI login.","This is especially important for:","When the access token expires, Codex starts returning auth failures instead of using the stored refresh token to recover automatically."],"workaround":null,"fixed_in":null,"score":292.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openai/codex#48237","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":48237,"title":"unexpected status 401 Unauthorized issue","url":"https://github.com/openai/codex/issues/48237","state":"open","state_reason":null,"labels":["bug","auth"],"created":"2026-09-25T22:54:35Z","updated":"2026-09-28T17:13:09Z","closed":null,"reactions":105,"thumbs_up":105,"comments":97,"severity":"high","severity_reasons":["engagement 202"],"topic":"auth/session","classes":["auth/session"],"cosmetic_ui":false,"summary":"unexpected status 401 Unauthorized: Incorrect API key provided: sk-svcacfvMA. You can find your API key at https://platform.openai.com/account/api-keys., url: https://chatgpt.com/backend-api/codex/responses, cf-ray: a40d95aa6bfcb8db-SEA…","versions":[],"error_snippets":[],"workaround":null,"fixed_in":null,"score":289.7,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openai/codex#48043","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":48043,"title":"[BUG] Codex CLI 0.157.0 fails to start on Windows with daemon privilege error (0.156.1 works)","url":"https://github.com/openai/codex/issues/48043","state":"open","state_reason":null,"labels":["bug","windows-os","CLI","app-server"],"created":"2026-09-25T09:34:59Z","updated":"2026-10-24T02:50:23Z","closed":null,"reactions":50,"thumbs_up":45,"comments":64,"severity":"critical","severity_reasons":["widespread start failure"],"topic":"always-on/daemon","classes":["always-on/daemon","self-host/headless","install/update"],"cosmetic_ui":false,"summary":"Codex CLI 0.157.0 fails to start on Windows with the following error: Downgrading to version 0.156.1 resolves the issue. Install Codex CLI 0.157.0 on Windows: Launch Codex CLI: Observe the daemon privilege error. Codex CLI should start…","versions":["0.157.0","0.156.1"],"error_snippets":["Error: start the Windows daemon from a non-elevated terminal; shared clients must not inherit administrator privileges","Codex CLI 0.157.0 fails to start on Windows with the following error:","3. Observe the daemon privilege error.","Codex CLI 0.157.0 fails to start and displays an error about launching the Windows daemon from a non-elevated terminal.","This appears to be a regression introduced in version 0.157.0, as the previous version works in the same environment."],"workaround":{"text":"Downgrading to version 0.156.1 restores normal operation:","source":"issue body","url":"https://github.com/openai/codex/issues/48043"},"fixed_in":null,"score":289.4,"first_seen":"2026-10-18T04:31:06Z"},{"id":"anthropics/claude-code#85891","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":85891,"title":"[BUG] Claude Desktop (Windows 11): main window stays always-on-top of other applications, with no setting to disable it","url":"https://github.com/anthropics/claude-code/issues/85891","state":"closed","state_reason":"completed","labels":["area:desktop"],"created":"2026-08-11T18:28:47Z","updated":"2026-10-23T06:59:56Z","closed":"2026-10-23T06:59:56Z","reactions":291,"thumbs_up":291,"comments":123,"severity":"high","severity_reasons":["engagement 414"],"topic":"always-on/daemon","classes":["always-on/daemon","install/update"],"cosmetic_ui":false,"summary":"On Windows 11, the Claude Desktop window behaves as a topmost window: it stays drawn above other applications even after another window is focused and brought over it. There is no in-app setting to turn this off. This is the Windows…","versions":[],"error_snippets":["On Windows 11, the Claude Desktop window behaves as a topmost window: it stays drawn above other applications even after another window is focused and brought over it. There is no in-app setting to turn this off.","2. Open another application (e.g. Outlook, Chrome, Excel, Bluebeam Revu) and position/move its window so it overlaps the Claude window.","Claude Desktop should behave like a normal Win32/Electron window: once another window is focused and overlaps it, Claude should be painted behind that window and follow standard z-order.","I checked before filing, in case something external was setting `WS_EX_TOPMOST`:","Either stop setting the always-on-top flag on the main window, or expose an explicit \"Keep window on top\" toggle in Settings that defaults to off."],"workaround":null,"fixed_in":null,"score":287.7,"first_seen":"2026-10-23T09:23:24Z"},{"id":"anthropics/claude-code#97074","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":97074,"title":"[BUG] Headless `claude -p` (entrypoint `sdk-cli`) uses ~1.8× more of the 5-hour window per token than interactive `cli`, same workload","url":"https://github.com/anthropics/claude-code/issues/97074","state":"open","state_reason":null,"labels":["bug","has repro","platform:linux","area:cost","area:cli","api:anthropic"],"created":"2026-09-25T11:00:35Z","updated":"2026-10-25T23:46:42Z","closed":null,"reactions":70,"thumbs_up":64,"comments":29,"severity":"medium","severity_reasons":["engagement 99","class:self-host/headless","class:auth/session"],"topic":"self-host/headless","classes":["self-host/headless","auth/session"],"cosmetic_ui":false,"summary":"On a Max subscription, the same workload uses about 1.8× more of the 5-hour window when it runs headless (claude -p --input-format stream-json, recorded as entrypoint: \"sdk-cli\" in the transcripts) than when it runs in the interactive TUI…","versions":["1.8","2.1.282","12.74","0.40"],"error_snippets":["Controlled A/B test, 2026-09-25 10:27–10:41 UTC, Claude Code 2.1.282, claude-opus-5, effort high","Phase | Test spend (API-equiv.) | Other spend | five_hour utilization | $ per 1% of 5h","Historical (same account, all Opus 5, API-equivalent spend per 1% of each 5h window that peaked above 25%):","entrypoint: \"sdk-cli\"","entrypoint: \"cli\""],"workaround":null,"fixed_in":null,"score":281.4,"first_seen":"2026-10-19T18:38:33Z"},{"id":"anthropics/claude-code#20571","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":20571,"title":"[Bug] VS Code extension: Tool use concurrency error after Jan 23-24 update","url":"https://github.com/anthropics/claude-code/issues/20571","state":"closed","state_reason":"completed","labels":["bug","has repro","platform:windows","area:api","area:ide","oncall"],"created":"2026-01-24T12:00:59Z","updated":"2026-10-13T15:43:06Z","closed":"2026-10-13T15:43:06Z","reactions":183,"thumbs_up":183,"comments":133,"severity":"high","severity_reasons":["engagement 316"],"topic":"install/update","classes":["install/update","mcp/tools"],"cosmetic_ui":false,"summary":"Bug Description VS Code extension update ~13 hours ago (Jan 23-24 2026) causes \"API Error: 400 due to tool use concurrency issues\" on every tool call. CLI works fine, only VS Code affected Error persists through extension reinstall, VS…","versions":["2.1.19"],"error_snippets":["VS Code extension update ~13 hours ago (Jan 23-24 2026) causes \"API Error: 400 due to tool use concurrency issues\" on every tool call.","Actual API error: \"messages.1: tool_use ids werefound without tool_result blocks immediately after\"","API Error: 400 due to tool use concurrency issues."],"workaround":{"text":"downgrade to vscode extension version 2.1.17 fixed the issue `uninstall the extension --> restart vscode --> install extension with 2.1.17 version -> fixed`","source":"comment by @QandilAFA (32 reactions)","url":"https://github.com/anthropics/claude-code/issues/20571#issuecomment-3794710623"},"fixed_in":null,"score":279.1,"first_seen":"2026-10-18T04:30:35Z"},{"id":"openai/codex#41290","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":41290,"title":"[Windows][WSL][26.825.31414] Project creation and removal fail after switching Agent Environment to WSL","url":"https://github.com/openai/codex/issues/41290","state":"open","state_reason":null,"labels":["bug","windows-os","app"],"created":"2026-08-28T06:39:10Z","updated":"2026-10-25T20:33:54Z","closed":null,"reactions":70,"thumbs_up":54,"comments":84,"severity":"high","severity_reasons":["engagement 154"],"topic":"self-host/headless","classes":["self-host/headless"],"cosmetic_ui":false,"summary":"26.825.31414 Microsoft Windows NT 10.0.26200.0 x64 ChatGPT / Powered by Codex & OWL Release date shown in About: August 28, 2026 Windows 11 x64 with WSL2 / Ubuntu. Project management works correctly when the Codex Agent Environment is set…","versions":["26.825.31414"],"error_snippets":["Agent Environment = Windows Native","Agent Environment = Windows Subsystem for Linux","Switch Agent Environment to WSL","Agent Environment = Windows Native → works","Agent Environment = WSL → fails"],"workaround":null,"fixed_in":null,"score":255.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"NousResearch/hermes-agent#18715","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":18715,"title":"Support remote Hermes agent with local tool execution","url":"https://github.com/NousResearch/hermes-agent/issues/18715","state":"open","state_reason":null,"labels":["type/feature","comp/agent","comp/gateway","comp/tools","tool/terminal","backend/local","area/config","P2","needs-decision"],"created":"2026-05-02T07:27:37Z","updated":"2026-10-25T12:36:17Z","closed":null,"reactions":44,"thumbs_up":42,"comments":31,"severity":"medium","severity_reasons":["engagement 75","class:always-on/daemon","class:self-host/headless"],"topic":"always-on/daemon","classes":["always-on/daemon","self-host/headless","mcp/tools"],"cosmetic_ui":false,"summary":"I want to use a remote Hermes Agent instance while keeping tool execution on my local machine. Machine A: local client / workspace machine Machine B: remote Hermes Agent with my existing skills, memory, sessions, and model/provider config…","versions":[],"error_snippets":["I want to use a remote Hermes Agent instance while keeping tool execution on my local machine.","This is surprising because normal LLM providers return `tool_calls` to the client, and the client executes tools locally.","For example:","Local machine should execute workspace-local tools:","Running everything on the remote Hermes host loses access to the local workspace, local services, local database, and local credentials."],"workaround":null,"fixed_in":null,"score":253.1,"first_seen":"2026-10-18T04:29:32Z"},{"id":"openai/codex#49458","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":49458,"title":"[Windows] dot-started local tasks lack Computer Use tools while ordinary local Codex sessions work","url":"https://github.com/openai/codex/issues/49458","state":"open","state_reason":null,"labels":["bug","windows-os","app","computer-use","remote","dots"],"created":"2026-09-30T03:09:53Z","updated":"2026-10-22T12:00:18Z","closed":null,"reactions":26,"thumbs_up":26,"comments":97,"severity":"medium","severity_reasons":["engagement 123","class:self-host/headless"],"topic":"self-host/headless","classes":["self-host/headless","install/update","auth/session","mcp/tools"],"cosmetic_ui":false,"summary":"ChatGPT desktop on Windows, with an About-dialog release date of September 29, 2026. The user confirmed the app is up to date. Installed Windows package version: 26.928.1915.0 (read from local package metadata, not the About-dialog display…","versions":["0.159.0","26.928.20755"],"error_snippets":[") followed by ready events with","failureReason=remote_unavailable","DesktopTaskWorkspaceUnavailableError","The exact About-dialog display version was not independently verified.","A dot can successfully start local tasks on my connected Windows computer, but those tasks cannot use Computer Use. An ordinary local Codex session on the same computer can inspect the current window using Computer Use."],"workaround":null,"fixed_in":null,"score":243.7,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openai/codex#48555","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":48555,"title":"[Android][Remote] \"Authorize this phone\" loops after the desktop switches ChatGPT accounts: stale cross-account environment, two pending enrollments per attempt","url":"https://github.com/openai/codex/issues/48555","state":"open","state_reason":null,"labels":["bug","auth","app","remote","Linux"],"created":"2026-09-26T22:18:20Z","updated":"2026-10-04T15:40:23Z","closed":null,"reactions":29,"thumbs_up":29,"comments":33,"severity":"medium","severity_reasons":["engagement 62","class:self-host/headless","class:auth/session"],"topic":"self-host/headless","classes":["self-host/headless","auth/session","install/update","network"],"cosmetic_ui":false,"summary":"After a desktop app was signed into account B and then back into account A on the same machine, pairing an Android phone with account A loops forever. After scanning the QR code, tapping Authorize this phone and completing the browser…","versions":["26.924.22138","0.158.0-alpha.2.1"],"error_snippets":["DELETE /backend-api/wham/remote/control/environments/{env_id}","GET .../wham/remote/control/environments","legacy_failure","1. With account B's credentials: `DELETE /backend-api/wham/remote/control/environments/{env_id}` for the stale environment whose `installation_id` matched the desktop now used by account A (HTTP 204). `GET .../wham/remot","When a desktop installation signs into a different account, the backend (or the app) should retire that installation's environment on the previous account, or the phone authorization should report an explicit account/ins"],"workaround":null,"fixed_in":null,"score":223.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openai/codex#45019","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":45019,"title":"App-server queued follow-up no longer exists","url":"https://github.com/openai/codex/issues/45019","state":"open","state_reason":null,"labels":["bug","app","app-server"],"created":"2026-09-12T11:08:51Z","updated":"2026-09-25T06:38:18Z","closed":null,"reactions":65,"thumbs_up":65,"comments":26,"severity":"medium","severity_reasons":["engagement 91","class:self-host/headless"],"topic":"self-host/headless","classes":["self-host/headless","auth/session"],"cosmetic_ui":false,"summary":"26.908.40834 Darwin 25.6.0 arm64 arm Trying to queue task after editing it once. So I had a queued task, then decided to edit it, and then I start to get this error. Submit queue, or submit queue edit it and try to submit again Queue being…","versions":["26.908.40834"],"error_snippets":["Error submitting message","App-server queued follow-up no longer exists","Trying to queue task after editing it once. So I had a queued task, then decided to edit it, and then I start to get this error.","Submit queue, or submit queue edit it and try to submit again","Queue being created normally or after edits"],"workaround":null,"fixed_in":null,"score":221.8,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openai/codex#44781","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":44781,"title":"[Codex Desktop] Editing and resending a queued message triggers \"App-server queued follow-up no longer exists\"","url":"https://github.com/openai/codex/issues/44781","state":"open","state_reason":null,"labels":["bug","windows-os","app","app-server"],"created":"2026-09-11T09:12:46Z","updated":"2026-09-30T10:17:36Z","closed":null,"reactions":54,"thumbs_up":54,"comments":48,"severity":"medium","severity_reasons":["engagement 102","class:self-host/headless"],"topic":"self-host/headless","classes":["self-host/headless","install/update"],"cosmetic_ui":false,"summary":"Installed Codex Desktop Windows package: 26.903.9818.0, verified locally with Get-AppxPackage. The executable used by the currently running Codex backend reports codex-cli 0.153.4 via --version. These were collected on September 11, 2026…","versions":["0.153.4"],"error_snippets":["App-server queued follow-up no longer exists","Editing and resending a queued follow-up message causes the app to display this error:","The Chinese heading means “Error creating chat.”","The reporter can reproduce this through the queued-message editing workflow. Switching to a different conversation thread in the app makes the error disappear.","3. Choose **Edit message** on that queued message."],"workaround":null,"fixed_in":null,"score":221.8,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openai/codex#48774","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":48774,"title":"Codex Remote pairing fails on Android","url":"https://github.com/openai/codex/issues/48774","state":"open","state_reason":null,"labels":["bug","windows-os","auth","app","remote"],"created":"2026-09-27T19:30:04Z","updated":"2026-10-08T03:24:25Z","closed":null,"reactions":27,"thumbs_up":27,"comments":55,"severity":"medium","severity_reasons":["engagement 82","class:self-host/headless","class:auth/session"],"topic":"self-host/headless","classes":["self-host/headless","auth/session"],"cosmetic_ui":false,"summary":"Codex Remote pairing fails on Android. ChatGPT mobile version 1.2026.265 and the Windows desktop app use the same verified personal account. “Allow connections” is enabled on the PC. After scanning a fresh QR code, the phone shows…","versions":["1.2026.265"],"error_snippets":[],"workaround":null,"fixed_in":null,"score":219.4,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openai/codex#46949","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":46949,"title":"Windows remote-control daemon spawns visible console windows for tool and MCP child processes","url":"https://github.com/openai/codex/issues/46949","state":"open","state_reason":null,"labels":["bug","windows-os","mcp","CLI","app-server","remote"],"created":"2026-09-21T06:22:47Z","updated":"2026-10-04T13:42:44Z","closed":null,"reactions":23,"thumbs_up":23,"comments":21,"severity":"medium","severity_reasons":["engagement 44","class:always-on/daemon","class:self-host/headless"],"topic":"always-on/daemon","classes":["always-on/daemon","self-host/headless","mcp/tools","network","install/update"],"cosmetic_ui":false,"summary":"On Windows, the managed remote-control app-server daemon launches tool-runtime and MCP child processes in visible console windows. Starting or opening a session through codex agents produces a cascade of black windows for noderepl.exe…","versions":["0.155.1"],"error_snippets":["5. Start or open an agent session with tool/MCP support enabled.","One configured stdio MCP makes the problem especially obvious:","Other visible window titles expose the bundled runtime paths for `node_repl.exe` and `node.exe`."],"workaround":null,"fixed_in":null,"score":215.5,"first_seen":"2026-10-18T04:34:23Z"},{"id":"anomalyco/opencode#2242","harness":"opencode","harness_name":"OpenCode","repo":"anomalyco/opencode","number":2242,"title":"Is there a way to sandbox the agent ?","url":"https://github.com/anomalyco/opencode/issues/2242","state":"open","state_reason":null,"labels":[],"created":"2025-08-25T15:08:23Z","updated":"2026-10-24T03:24:15Z","closed":null,"reactions":87,"thumbs_up":80,"comments":103,"severity":"high","severity_reasons":["engagement 190"],"topic":"other","classes":[],"cosmetic_ui":false,"summary":"Is there a way to restrict terminal command from a agent to access/edit file outside the current directory ? I see that gemini-cli or codex-cli use seatbelt on macOS. I don't see any equivalent for copencode.","versions":[],"error_snippets":["Is there a way to restrict terminal command from a agent to access/edit file outside the current directory ?","I don't see any equivalent for copencode."],"workaround":null,"fixed_in":null,"score":212.1,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openai/codex#48333","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":48333,"title":"[Windows] Codex Desktop 26.924.1866.0 stuck on startup spinner until app-server codex.exe is terminated","url":"https://github.com/openai/codex/issues/48333","state":"open","state_reason":null,"labels":["bug","windows-os","mcp","app","app-server"],"created":"2026-09-26T02:18:57Z","updated":"2026-10-02T06:34:52Z","closed":null,"reactions":16,"thumbs_up":13,"comments":29,"severity":"medium","severity_reasons":["engagement 45","class:self-host/headless","class:crash/hang"],"topic":"self-host/headless","classes":["self-host/headless","crash/hang","mcp/tools","install/update","auth/session"],"cosmetic_ui":false,"summary":"26.924.1866.0 (Windows package version; the app does not load reliably enough to open About Codex) ChatGPT Plus Microsoft Windows NT 10.0.19045.0 x64 Codex Desktop remains indefinitely on the startup spinner before showing projects or…","versions":[],"error_snippets":["26.924.1866.0 (Windows package version; the app does not load reliably enough to open About Codex)","Codex Desktop remains indefinitely on the startup spinner before showing projects or chats. Its child codex.exe app-server process stays running. If I terminate only that process while the spinner is visible, the UI proc","MCP server startup failed server_name=\"codex_app\" error=\"timed out awaiting tools/list after 10s\"","5. The UI advances and displays the existing chats.","codex.exe -c features.code_mode_host=true app-server --analytics-default-enabled -c plugins.codex-app-tools@openai-bundled.mcp_servers.codex_app.enabled=true"],"workaround":null,"fixed_in":null,"score":211.7,"first_seen":"2026-10-18T04:34:23Z"},{"id":"openai/codex#51601","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":51601,"title":"Windows app 26.1002.51308: sandbox setup fails with sharing violation when validating its own active runtime","url":"https://github.com/openai/codex/issues/51601","state":"open","state_reason":null,"labels":["bug","windows-os","sandbox","app"],"created":"2026-10-07T06:33:31Z","updated":"2026-10-15T03:45:58Z","closed":null,"reactions":31,"thumbs_up":31,"comments":124,"severity":"high","severity_reasons":["engagement 155"],"topic":"install/update","classes":["install/update","network"],"cosmetic_ui":false,"summary":"Title: Windows app 26.1002.51308: sandbox setup fails with sharing violation when validating its own active runtime After updating the Windows Codex app to 26.1002.51308 (build 13417), all command execution fails before the requested…","versions":["26.1002.51308","0.160.1"],"error_snippets":["helper_unknown_error: setup refresh had errors","trusted Node process exited unexpectedly","runtime read/execute validation failed: validate runtime read/execute access on","open ACL target for root-only update: The process cannot access the file because it is being used by another process. (os error 32)","ERROR_SHARING_VIOLATION"],"workaround":null,"fixed_in":null,"score":211.4,"first_seen":"2026-10-18T04:34:23Z"},{"id":"anthropics/claude-code#102150","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":102150,"title":"[BUG] Desktop 2.37337.1 (Windows, Store/MSIX): built-in \"Routines\" sidebar section is forced hidden on every launch, and there is no UI toggle to show it","url":"https://github.com/anthropics/claude-code/issues/102150","state":"open","state_reason":null,"labels":["bug","has repro","platform:windows","regression","area:ui","area:desktop","area:routines"],"created":"2026-10-14T19:17:25Z","updated":"2026-10-23T20:36:29Z","closed":null,"reactions":34,"thumbs_up":34,"comments":21,"severity":"high","severity_reasons":["label:regression"],"topic":"install/update","classes":["install/update","always-on/daemon","self-host/headless"],"cosmetic_ui":false,"summary":"Claude Desktop 2.37337.1, Windows Store/MSIX package (Claudepzs8sxrjxfjjc), self-updated on 2026-10-13 15:11 local time Windows 11 Pro 10.0.26200 Code tab, sidebar in sections layout (sidebarMode: epitaxy) After the update, the dedicated…","versions":["10.0.26200"],"error_snippets":["sidebarMode: epitaxy","preferences.epitaxyPrefs[\"dframe-code-sections\"][<account/org>].sections[kind == \"routines\"]","After the update, the dedicated **Routines** section disappeared from the Code-tab sidebar. It used to list my four routines (three local scheduled tasks: a weekday 7:59 am morning brief, a nightly Dependabot sweep, a Su","The section still exists in the app's sidebar model, but flagged hidden. The app's own `list_sections` tool reports:","2. **Computer use.** Cannot be granted on Claude's own window (by design), so no agent can click the toggle even if one existed."],"workaround":null,"fixed_in":null,"score":208.1,"first_seen":"2026-10-18T04:33:44Z"},{"id":"openai/codex#48777","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":48777,"title":"Android 1.2026.265 (27): Codex Remote repeatedly returns to “Authorize this phone” after successful browser authorization","url":"https://github.com/openai/codex/issues/48777","state":"open","state_reason":null,"labels":["bug","auth","CLI","remote"],"created":"2026-09-27T19:52:49Z","updated":"2026-10-21T16:06:00Z","closed":null,"reactions":21,"thumbs_up":21,"comments":38,"severity":"medium","severity_reasons":["engagement 59","class:self-host/headless","class:auth/session"],"topic":"self-host/headless","classes":["self-host/headless","auth/session","always-on/daemon"],"cosmetic_ui":false,"summary":"I cannot finish pairing my Android phone with Codex running in WSL. The phone opens the browser login, recognizes the correct ChatGPT account, and asks me to authorize access. After I authorize, it briefly shows “Pairing…” and returns to…","versions":["1.2026.265","0.157.1"],"error_snippets":["I cannot finish pairing my Android phone with Codex running in WSL. The phone opens the browser login, recognizes the correct ChatGPT account, and asks me to authorize access. After I authorize, it briefly shows “Pairing","My environment:","I expect authorization to complete and the phone to finish pairing with the computer. Instead, the app repeatedly sends me through the login and authorization flow without completing the connection.","I have not established whether my failure is in the Android app, the authorization callback, or backend device enrollment. Similar reports on other host operating systems suggest that WSL may not be the cause.","Please investigate the Android Remote authorization and device-enrollment flow for **1.2026.265 (27)**. Is there a confirmed fixed build or a supported recovery procedure? Please also specify which sanitized diagnostics "],"workaround":null,"fixed_in":null,"score":205.2,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openai/codex#48500","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":48500,"title":"Managed app-server runs hooks with the first client's TMUX_PANE/TMUX, misattributing hook events to the wrong terminal pane (0.157 regression)","url":"https://github.com/openai/codex/issues/48500","state":"open","state_reason":null,"labels":["bug","CLI","hooks","app-server"],"created":"2026-09-26T18:00:55Z","updated":"2026-10-23T18:11:47Z","closed":null,"reactions":23,"thumbs_up":23,"comments":19,"severity":"high","severity_reasons":["title:breakage"],"topic":"self-host/headless","classes":["self-host/headless","install/update","always-on/daemon"],"cosmetic_ui":false,"summary":"Since 0.157 the TUI talks to a shared codex app-server --managed-daemon (one per CODEXHOME), and lifecycle hooks (UserPromptSubmit, Stop, PostToolUse) run inside that daemon. The daemon inherits the environment of whichever TUI first…","versions":["0.157","0.157.1"],"error_snippets":[") run inside that daemon. The daemon inherits the environment of whichever TUI first spawned it, so every hook for every later TUI sees that first client's","Tmux-aware hook consumers (e.g. workmux status tracking) silently write B's session to A's pane (input/status then route to the wrong session), or drop B's updates entirely."],"workaround":null,"fixed_in":null,"score":188.7,"first_seen":"2026-10-18T04:34:23Z"},{"id":"NousResearch/hermes-agent#143547","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":143547,"title":"Unified gateway (#106742): open post-merge problems on main, continuing #134889","url":"https://github.com/NousResearch/hermes-agent/issues/143547","state":"open","state_reason":null,"labels":["comp/gateway","comp/cron","area/auth","P2","comp/desktop","area/sessions","area/install-update","area/profiles","desktop/sessions","os/windows","comp/tui-gateway","desktop/connection"],"created":"2026-10-23T13:00:12Z","updated":"2026-10-26T09:54:01Z","closed":null,"reactions":1,"thumbs_up":1,"comments":23,"severity":"medium","severity_reasons":["engagement 24","class:always-on/daemon","class:install/update"],"topic":"always-on/daemon","classes":["always-on/daemon","install/update","auth/session","self-host/headless","crash/hang","network"],"cosmetic_ui":false,"summary":"What changed. On 14 October, #106742 (the unified gateway) moved every local conversation into one gateway process. The CLI, the terminal UI, the local Desktop app, the API server, editors, messaging bots, cron jobs, Kanban cards and Group…","versions":[],"error_snippets":["first. A row that doesn't reproduce gets no commit: post the evidence (","hermes gateway ensure","); the CLI edits the same rows. #142942 (@davidmeischner99), #143110 (@clientamp) | broken | 🟢 fixed on","PermissionError","is never read. #138553 (@markdebont) (older code; reached by everyone after #106742's multiplex migration) | broken | 🟢 fixed on"],"workaround":{"text":"The only workaround is config (`ws_orphan_reap_grace_s=0` plus a higher `HERMES_TUI_SESSION_TTL_S`). \"I expected orchestration to continue with the laptop off.\" - [https://discord.com/channels/1053877538025386074/1563120848624164905](https://","source":"comment by @teknium1","url":"https://github.com/NousResearch/hermes-agent/issues/143547#issuecomment-6399971267"},"fixed_in":null,"score":179.8,"first_seen":"2026-10-24T08:54:58Z"},{"id":"openai/codex#49729","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":49729,"title":"Dot cannot create or follow up with local Codex tasks in saved projects","url":"https://github.com/openai/codex/issues/49729","state":"open","state_reason":null,"labels":["bug","app","app-server","dots"],"created":"2026-09-30T20:11:09Z","updated":"2026-10-25T11:52:03Z","closed":null,"reactions":12,"thumbs_up":12,"comments":61,"severity":"medium","severity_reasons":["engagement 73","class:self-host/headless"],"topic":"self-host/headless","classes":["self-host/headless"],"cosmetic_ui":false,"summary":"A dot can start a local Codex task on a connected desktop, but its task-creation tool cannot select an existing saved Codex App project. A local Codex task can create a thread in that project, yet the dot then cannot read or message that…","versions":[],"error_snippets":["CloudThreadNotFoundError","A dot can start a local Codex task on a connected desktop, but its task-creation tool cannot select an existing saved Codex App project. A local Codex task can create a thread in that project, yet the dot then cannot rea","1. Connect a Mac running ChatGPT/Codex Desktop to a dot and register an existing local Git repository as a Codex project.","3. Observe that the dot's task creation interface accepts a computer and `cwd`, but exposes no project target. A task it creates directly appears projectless.","The project is the user's normal place to manage related coding work. A working-directory-only task may load repository `AGENTS.md`, but it is displayed outside the project and cannot participate in the expected project "],"workaround":null,"fixed_in":null,"score":171.4,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openai/codex#48189","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":48189,"title":"[Linux] Codex Desktop 26.924.20706 hangs indefinitely on \"Starting your task\"; rollback to 26.917.71314 fixes it","url":"https://github.com/openai/codex/issues/48189","state":"closed","state_reason":"completed","labels":["bug","app","app-server"],"created":"2026-09-25T19:46:48Z","updated":"2026-09-28T21:16:38Z","closed":"2026-09-28T21:16:38Z","reactions":66,"thumbs_up":60,"comments":33,"severity":"high","severity_reasons":["title:breakage"],"topic":"self-host/headless","classes":["self-host/headless","crash/hang","install/update"],"cosmetic_ui":false,"summary":"26.924.20706 Linux Mint, X11, amd64 After updating from 26.917.71314 to 26.924.20706, every local Codex task hangs indefinitely at \"Starting your task\". This reproduces with both an existing project task and a brand-new trivial task…","versions":["26.924.20706","26.917.71314"],"error_snippets":["After updating from 26.917.71314 to 26.924.20706, every local Codex task hangs indefinitely at \"Starting your task\".","This reproduces with both an existing project task and a brand-new trivial task.","\"ignoring invalid experimental feature enablement keys: apps_mcp_path_override\""],"workaround":{"text":"Downgrade to 26.917.71314.","source":"issue body","url":"https://github.com/openai/codex/issues/48189"},"fixed_in":null,"score":167.8,"first_seen":"2026-10-18T04:31:06Z"},{"id":"anthropics/claude-code#102136","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":102136,"title":"[BUG] Permission prompt defaults to \"Always allow\" instead of \"Allow once\" - dangerous UX regression","url":"https://github.com/anthropics/claude-code/issues/102136","state":"open","state_reason":null,"labels":["bug","platform:macos","user-experience","regression","area:permissions","area:desktop"],"created":"2026-10-14T18:29:54Z","updated":"2026-10-23T18:45:38Z","closed":null,"reactions":22,"thumbs_up":22,"comments":14,"severity":"high","severity_reasons":["label:regression","title:breakage"],"topic":"install/update","classes":["install/update","network"],"cosmetic_ui":false,"summary":"This has been reported but was closed for being stale. Reopening in a new ticket. The most recent upgrade to Claude Code regressed on https://github.com/anthropics/claude-code/issues/48550 which was auto closed for inactivity. I cannot…","versions":[],"error_snippets":["`I have searched existing issues and this hasn't been reported yet`","The most recent upgrade to Claude Code regressed on https://github.com/anthropics/claude-code/issues/48550 which was auto closed for inactivity. I cannot allow claude to start running python whenever it wants.","The permission prompt in Claude Code Desktop now defaults to \"Always allow\" instead of \"Allow once\". Users reflexively pressing Enter silently grant permanent permissions they never intended to give.","Clear Regression, Allow once shouldn't be multiple clicks and shouldn't be hidden in a menu for high risk commands especially."],"workaround":null,"fixed_in":null,"score":164.2,"first_seen":"2026-10-18T04:33:44Z"},{"id":"openai/codex#56691","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":56691,"title":"[BUG] Codex 26.1021.22027 fails to start UI in VS Code 1.143.1 on Windows","url":"https://github.com/openai/codex/issues/56691","state":"open","state_reason":null,"labels":["bug","windows-os","extension"],"created":"2026-10-24T09:46:40Z","updated":"2026-10-26T07:49:37Z","closed":null,"reactions":19,"thumbs_up":19,"comments":31,"severity":"critical","severity_reasons":["widespread start failure"],"topic":"always-on/daemon","classes":["always-on/daemon","install/update","auth/session"],"cosmetic_ui":false,"summary":"26.1021.22027 ChatGPT Plus Visual Studio Code 1.143.1 (Windows) Microsoft Windows NT 10.0.19045.0 x64 Codex IDE extension version 26.1021.22027 fails to initialize its user interface in Visual Studio Code 1.143.1 on Windows. When opening…","versions":["26.1021.22027","1.143.1","26.1007.21434"],"error_snippets":["The extension could not start its user interface.","Error loading webview provided by 'openai.chatgpt':","InvalidStateError: Failed to register a ServiceWorker:","InvalidStateError","code --disable-gpu --user-data-dir \"$env:TEMP\\vscode-codex-isolado\" --extensions-dir \"$env:TEMP\\vscode-ext-isoladas\""],"workaround":{"text":"Confirmed workaround: Downgrading the Codex extension from `26.1021.22027` to `26.1007.21434` immediately resolved the issue.","source":"issue body","url":"https://github.com/openai/codex/issues/56691"},"fixed_in":null,"score":164.1,"first_seen":"2026-10-26T10:12:02Z"},{"id":"omacom/omarchy#5105","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":5105,"title":"Update/reboot left keyring in invalid state, browser sessions and gh auth got cleared","url":"https://github.com/omacom/omarchy/issues/5105","state":"open","state_reason":null,"labels":["bug"],"created":"2026-03-23T18:06:58Z","updated":"2026-10-25T14:29:36Z","closed":null,"reactions":11,"thumbs_up":11,"comments":29,"severity":"medium","severity_reasons":["engagement 40","class:install/update","class:auth/session"],"topic":"install/update","classes":["install/update","auth/session"],"cosmetic_ui":false,"summary":"On Omarchy master, Hyprland 0.54.2, kernel 6.19.8-arch1-1 After updating Omarchy and rebooting my laptop, my browser logins/sessions were cleared and my GitHub CLI auth was also gone I have not fully diagnosed the root cause, but this…","versions":["0.54.2"],"error_snippets":["Possible keyring evidence:"],"workaround":{"text":"I applied the documented workaround: blanked the `Default` keyring password via seahorse.","source":"comment by @jparlay","url":"https://github.com/omacom/omarchy/issues/5105#issuecomment-6251287775"},"fixed_in":null,"score":162.7,"first_seen":"2026-10-19T18:38:33Z"},{"id":"anthropics/claude-code#94393","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":94393,"title":"[BUG] Monitor tool: doesn't respect persistent flag and timeout_ms schema caps at 3600000 but actual lifetime is ~30 minutes, even during active (non-idle) sessions","url":"https://github.com/anthropics/claude-code/issues/94393","state":"open","state_reason":null,"labels":["bug","has repro","platform:linux","area:tools"],"created":"2026-09-14T21:51:59Z","updated":"2026-10-23T11:07:03Z","closed":null,"reactions":30,"thumbs_up":30,"comments":13,"severity":"medium","severity_reasons":["engagement 43","class:network"],"topic":"network","classes":["network","always-on/daemon"],"cosmetic_ui":false,"summary":"The Monitor tool now hard-rejects persistent and timeoutms above 3600000 (1 hour) even with persistent set, it rejects with a schema validation error, and even a Monitor started with the maximum now allowed (3600000) is reported as…","versions":[],"error_snippets":["InputValidationError: [","\"path\": [\"timeout_ms\"],","\"message\": \"timeout_ms must be <= 3600000\"","Monitor started (task <id>, expires in 30m unless the source ends first; you get one notice at expiry — re-arm if you still need the watch)","Monitor started (task <id>, expires in 30m unless the source ends first; you get one notice at expiry — re-arm if you still need the watch). You will be notified on each event. Keep working — do not poll or sleep. Events"],"workaround":null,"fixed_in":null,"score":161.7,"first_seen":"2026-10-18T04:33:44Z"},{"id":"openai/codex#51763","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":51763,"title":"Background daemon auto-update drains active session and leaves daemon incompatible with current Codex client","url":"https://github.com/openai/codex/issues/51763","state":"open","state_reason":null,"labels":["bug"],"created":"2026-10-07T16:52:34Z","updated":"2026-10-08T04:17:51Z","closed":null,"reactions":20,"thumbs_up":20,"comments":1,"severity":"medium","severity_reasons":["engagement 21","class:always-on/daemon","class:install/update"],"topic":"always-on/daemon","classes":["always-on/daemon","install/update","self-host/headless","crash/hang","auth/session"],"cosmetic_ui":false,"summary":"An active Codex session suddenly failed with: I did not update Codex or change feature settings while the session was running. Afterwards, attempting to resume the session produced: Importantly, this is not specific to the crashed session…","versions":["0.160.1","0.161.0"],"error_snippets":["Error: turn/steer failed: Server is draining; retry after reconnecting (code -32600)","feature flags: 53 enabled · 1 overridden","mode: ephemeral","enabled = true","An active Codex session suddenly failed with:"],"workaround":null,"fixed_in":null,"score":161.0,"first_seen":"2026-10-18T04:34:23Z"},{"id":"openai/codex#52483","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":52483,"title":"[macOS] “More details” missing from response selection menu after update to 26.1007.21159","url":"https://github.com/openai/codex/issues/52483","state":"open","state_reason":null,"labels":["bug","app"],"created":"2026-10-09T08:53:31Z","updated":"2026-10-23T21:48:09Z","closed":null,"reactions":21,"thumbs_up":20,"comments":19,"severity":"medium","severity_reasons":["engagement 40","class:install/update"],"topic":"install/update","classes":["install/update","crash/hang"],"cosmetic_ui":false,"summary":"ChatGPT desktop 26.1007.21159, build 20052. The app’s updater reports that this version is current. ChatGPT Pro. macOS 26.6.2, build 25G83. After the latest update, the contextual menu shown when selecting assistant response text no longer…","versions":["26.1007.21159"],"error_snippets":["The menu itself appears, and **Add to chat** and **Ask in side chat** are still available. The missing action is specifically **More details**, which previously opened a Quick Chat window to explain the selected passage ","Resetting app zoom with Command–0 and comparing the selection at normal and enlarged zoom did not restore the action.","I have not established the cause or the exact first affected version. I have not tested a rollback.","I checked existing reports before filing this. #25641 concerns the entire selection toolbar failing to appear, and #37520 concerns the missing **Ask in side chat** action; neither describes this symptom. My earlier comme"],"workaround":null,"fixed_in":null,"score":160.9,"first_seen":"2026-10-26T10:12:02Z"},{"id":"anthropics/claude-code#101685","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":101685,"title":"[BUG] Interactive sessions write the 5-minute prompt cache despite requesting 1h: `cache-keepalive-2026-09-03` beta (2.1.296–2.1.300, Team and Max; `claude -p` unaffected)","url":"https://github.com/anthropics/claude-code/issues/101685","state":"open","state_reason":null,"labels":["bug","area:core","platform:wsl"],"created":"2026-10-13T05:30:19Z","updated":"2026-10-20T09:07:14Z","closed":null,"reactions":12,"thumbs_up":12,"comments":11,"severity":"medium","severity_reasons":["engagement 23","class:always-on/daemon","class:self-host/headless"],"topic":"always-on/daemon","classes":["always-on/daemon","self-host/headless","install/update","auth/session"],"cosmetic_ui":false,"summary":"Since 2026-10-12 at about 18:18 UTC, every newly started interactive session writes its main-conversation prompt cache under the 5-minute TTL. Headless claude -p runs on the same account, machine, version and model still write under 1…","versions":["2.1.296","2.1.300"],"error_snippets":["| [.timestamp, .entrypoint, .message.usage.cache_creation.ephemeral_1h_input_tokens,","fallback_credit: null","worked on the same account. One investigator traced it to per-session flag evaluation, and one report involved","1. Run `claude` interactively, send one turn and exit."],"workaround":null,"fixed_in":null,"score":156.7,"first_seen":"2026-10-19T18:38:33Z"},{"id":"github/copilot-cli#4095","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":4095,"title":"Windows: plugin update fails with \"Access is denied (os error 5)\" while VS Code is running (Copilot extension holds watcher handles on installed-plugins)","url":"https://github.com/github/copilot-cli/issues/4095","state":"open","state_reason":null,"labels":["area:platform-windows","area:plugins"],"created":"2026-07-11T19:18:47Z","updated":"2026-09-17T10:59:18Z","closed":null,"reactions":25,"thumbs_up":25,"comments":3,"severity":"high","severity_reasons":["title:breakage"],"topic":"install/update","classes":["install/update","mcp/tools"],"cosmetic_ui":false,"summary":"copilot plugin update (and the same flow triggered from the GitHub Copilot desktop app) fails on Windows with: The git fetch/checkout of the marketplace clone succeeds every time (the clone under %LOCALAPPDATA%\\copilot\\marketplaces\\…","versions":["2.5.0","3.0.0","1.0.70","1.0.21"],"error_snippets":["Failed to update plugin: Error: Failed to install plugin: Error: Access is denied. (os error 5)","ERROR_ACCESS_DENIED (5)","Failed to install plugin: Error: Access is denied. (os error 5)","The git fetch/checkout of the marketplace clone succeeds every time (the clone under `%LOCALAPPDATA%\\copilot\\marketplaces\\...` ends up at the target commit, working tree clean). The failure happens in the final step that","Root cause (verified with a system-wide handle enumeration via `NtQuerySystemInformation(SystemExtendedHandleInformation)`): **VS Code's Copilot extension holds directory watcher handles** (access mask `0x100081`, i.e. `"],"workaround":null,"fixed_in":null,"score":151.6,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openai/codex#51590","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":51590,"title":"Windows sandbox fails opening running node_repl.exe for ACL update (error 32); Computer Use and shell blocked","url":"https://github.com/openai/codex/issues/51590","state":"open","state_reason":null,"labels":["bug","windows-os","sandbox","app","computer-use"],"created":"2026-10-07T05:36:39Z","updated":"2026-10-16T21:38:40Z","closed":null,"reactions":11,"thumbs_up":11,"comments":40,"severity":"medium","severity_reasons":["engagement 51","class:install/update"],"topic":"install/update","classes":["install/update","network"],"cosmetic_ui":false,"summary":"On Windows 11, Computer Use cannot initialize and ordinary sandboxed shell commands fail before execution with helperunknownerror: setup refresh had errors. The sandbox log identifies an error 32 (sharing violation) when opening the…","versions":["0.162.0-alpha.2","24.21.0-cua.1","24.21.0"],"error_snippets":["node_repl kernel exited unexpectedly","windows sandbox failed: helper_unknown_error: setup refresh had errors","Failed to create unified exec process:","helper_unknown_error: setup refresh had errors","runtime read/execute validation failed:"],"workaround":null,"fixed_in":null,"score":150.8,"first_seen":"2026-10-18T04:34:23Z"},{"id":"github/copilot-cli#5068","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":5068,"title":"Windows: MCP Entra sign-in fails with \"this server's advertised scopes could not be safely validated for the account broker\"","url":"https://github.com/github/copilot-cli/issues/5068","state":"open","state_reason":null,"labels":["area:platform-windows","area:mcp"],"created":"2026-10-06T22:44:28Z","updated":"2026-10-12T12:38:07Z","closed":null,"reactions":12,"thumbs_up":12,"comments":3,"severity":"medium","severity_reasons":["engagement 15","class:self-host/headless","class:auth/session"],"topic":"self-host/headless","classes":["self-host/headless","auth/session","mcp/tools","crash/hang"],"cosmetic_ui":false,"summary":"On Windows, authenticating to an Entra ID–protected MCP server (Microsoft's hosted Azure DevOps MCP server, https://mcp.dev.azure.com/{org} , server type http ) fails on every fresh interactive sign-in attempt with: Authentication failed…","versions":["1.0.93-2","1.0.92","1.0.81"],"error_snippets":["On Windows, authenticating to an Entra ID–protected MCP server (Microsoft's hosted Azure DevOps MCP server, https://mcp.dev.azure.com/{org} , server type http ) fails on every fresh interactive sign-in attempt with:","Authentication failed: MCPOAuthError: Microsoft Entra sign-in for https://mcp.dev.azure.com/{org} failed: this server's advertised scopes could not be safely validated for the account broker","/mcp auth does not open the login/account-picker dialog at all — it fails immediately with the above error.","4. Expected: Windows account picker / browser OAuth dialog opens for Entra sign-in.","5. Actual: Dialog never appears; command fails instantly with the \"advertised scopes could not be safely validated for the account broker\" error."],"workaround":null,"fixed_in":null,"score":147.8,"first_seen":"2026-10-19T18:38:33Z"},{"id":"anthropics/claude-code#95364","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":95364,"title":"Desktop auto-update (\"stealth update\") quits and relaunches the app while the user is away, dropping every Remote Control session","url":"https://github.com/anthropics/claude-code/issues/95364","state":"open","state_reason":null,"labels":["bug","has repro","platform:macos","area:desktop"],"created":"2026-09-18T13:07:40Z","updated":"2026-10-23T03:23:29Z","closed":null,"reactions":14,"thumbs_up":14,"comments":12,"severity":"medium","severity_reasons":["engagement 26","class:self-host/headless","class:install/update"],"topic":"self-host/headless","classes":["self-host/headless","install/update","network"],"cosmetic_ui":false,"summary":"Claude Desktop 2.2553.1 (macOS 26.6.2, Apple Silicon) Bundled Claude Code 2.1.275 Remote Control in use daily from claude.ai/code while away from the machine When an update is staged, the app waits for a short idle window and then quits…","versions":["2.2553.1","2.1.275","2.2553.0"],"error_snippets":["10:21:28 [info] [stealth-update] Triggering stealth update after idle timeout","10:21:30 [info] [CCD] CLI exit before update: 3 of 3 exited within 1772ms","When an update is staged, the app waits for a short idle window and then **quits and relaunches itself**. It logs this as `[stealth-update]`. The relaunch stops every running Code session and drops each session's Remote ","For a user sitting at the machine this is nearly invisible, which is presumably the intent. For a user working **remotely** it is the opposite of invisible: the restart fires precisely when nobody is at the keyboard, whi","1. the same version having failed to install repeatedly"],"workaround":null,"fixed_in":null,"score":147.4,"first_seen":"2026-10-18T04:33:44Z"},{"id":"anthropics/claude-code#102088","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":102088,"title":"[BUG] Desktop 2.37337.1: “Group by” option removed from Code tab sidebar — sessions now only grouped by environment","url":"https://github.com/anthropics/claude-code/issues/102088","state":"closed","state_reason":"completed","labels":["bug","platform:windows","regression","area:ui","area:desktop"],"created":"2026-10-14T16:16:04Z","updated":"2026-10-19T18:06:33Z","closed":"2026-10-19T18:06:33Z","reactions":46,"thumbs_up":46,"comments":63,"severity":"high","severity_reasons":["label:regression"],"topic":"install/update","classes":["install/update","self-host/headless"],"cosmetic_ui":false,"summary":"Claude desktop app: 2.37337.1 Code tab sidebar After a recent update, the Code tab sidebar no longer groups sessions by project/repository. Sessions are now grouped only by environment: \"This computer\", \"Cloud\", SSH host (e.g. ubuntu@ )…","versions":[],"error_snippets":["No error messages. The option is simply missing from the UI.","project/repository. Sessions are now grouped only by environment:","scattered across different environment groups (local / cloud / SSH),","3. Look at the session sidebar: sessions are grouped by environment","5. There is no \"Group by\" option, so project grouping cannot be restored"],"workaround":{"text":"I had Claude create a custom sidebar group for each of my 15 project folders and move ~127 sessions into them.","source":"comment by @lexxenbourg (5 reactions)","url":"https://github.com/anthropics/claude-code/issues/102088#issuecomment-6174881673"},"fixed_in":null,"score":145.3,"first_seen":"2026-10-18T04:33:44Z"},{"id":"anthropics/claude-code#103376","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":103376,"title":"[BUG] Weekly limit 0% to 54% in 22h with same token volume that took 4.4 days to reach 60% the prior week","url":"https://github.com/anthropics/claude-code/issues/103376","state":"open","state_reason":null,"labels":["bug","platform:macos","area:cost"],"created":"2026-10-19T06:30:55Z","updated":"2026-10-25T07:09:28Z","closed":null,"reactions":12,"thumbs_up":12,"comments":9,"severity":"medium","severity_reasons":["engagement 21","class:auth/session"],"topic":"auth/session","classes":["auth/session","always-on/daemon","self-host/headless","install/update"],"cosmetic_ui":false,"summary":"On Claude Max (20x), the weekly \"all models\" meter went from 0% to 54% in the first 22 hours after the weekly reset. My Claude Code token volume in those 22 hours was nearly identical to the same 22 hours of the previous week. The previous…","versions":["4.4","2.1.299","2.1.303"],"error_snippets":["\"api_error\":\"usage_limit_reached\",\"api_error_status\":429,\"result\":\"You've hit your session limit\"","Token counts come from my local Claude Code transcripts (`~/.claude/projects/**/*.jsonl`, deduplicated by message ID and request ID, ` ` excluded):","The only other Claude Code activity on the account is one Remote Control session on my second computer. Its event log shows *less* use this Sunday than the previous Sunday (212 calls and 44.9M cache read, against 697 cal","The same token volume should consume about the same share of the weekly limit from week to week. At the prior week's rate, everything I can measure this week should come to roughly 9 to 13%, not 54%.","Remote Control session result on Sunday night: `\"api_error\":\"usage_limit_reached\",\"api_error_status\":429,\"result\":\"You've hit your session limit\"`"],"workaround":null,"fixed_in":null,"score":144.1,"first_seen":"2026-10-22T09:27:06Z"},{"id":"openai/codex#48232","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":48232,"title":"Remote compaction fails with 401 'Incorrect API key' (sk-svcacct) for ChatGPT-authenticated users","url":"https://github.com/openai/codex/issues/48232","state":"closed","state_reason":"duplicate","labels":["bug","auth","CLI","context"],"created":"2026-09-25T22:48:16Z","updated":"2026-09-25T23:35:53Z","closed":"2026-09-25T23:07:46Z","reactions":44,"thumbs_up":42,"comments":32,"severity":"medium","severity_reasons":["engagement 76","class:self-host/headless","class:auth/session"],"topic":"self-host/headless","classes":["self-host/headless","auth/session","always-on/daemon"],"cosmetic_ui":false,"summary":"Remote compaction fails with a 401. The rejected key is an OpenAI service-account key (sk-svcacct-…), but that key isn't configured anywhere on the client. The session shows \"Compaction interrupted\" followed by \"Reconnecting 2/…\". The…","versions":["0.157.0"],"error_snippets":["Error running remote compact task: unexpected status 401 Unauthorized: Incorrect API key provided: sk-svcac***…fvMA. You can find your API key at https://platform.openai.com/account/api-keys., url: https://chatgpt.com/ba","Remote compaction fails with a 401. The rejected key is an OpenAI **service-account key** (`sk-svcacct-…`), but that key isn't configured anywhere on the client. The session shows \"Compaction interrupted\" followed by \"Re"],"workaround":null,"fixed_in":null,"score":144.1,"first_seen":"2026-10-19T18:38:33Z"},{"id":"anthropics/claude-code#95967","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":95967,"title":"Scheduled task (routine) runs no longer appear in the desktop sidebar or Remote Control","url":"https://github.com/anthropics/claude-code/issues/95967","state":"open","state_reason":null,"labels":["bug","has repro","platform:macos","area:desktop","area:routines"],"created":"2026-09-22T04:13:52Z","updated":"2026-10-21T07:08:46Z","closed":null,"reactions":13,"thumbs_up":13,"comments":19,"severity":"medium","severity_reasons":["engagement 32","class:always-on/daemon","class:self-host/headless"],"topic":"always-on/daemon","classes":["always-on/daemon","self-host/headless"],"cosmetic_ui":false,"summary":"What I expect I run local scheduled tasks (routines) in Claude Desktop on my Mac. Each run used to show up as a normal session: in the desktop app's left sidebar, and in Remote Control on my phone, so I could open a run and continue it…","versions":["2.1.278","2.2553.1"],"error_snippets":["I run local scheduled tasks (routines) in Claude Desktop on my Mac. Each run used to show up as a normal session: in the desktop app's left sidebar, and in Remote Control on my phone, so I could open a run and continue i","Routine runs no longer appear in the sidebar or in Remote Control. I can't open or continue one from my phone or the sidebar anymore.","So the app itself has already marked the run as eligible for Remote Control and not archived — but it still doesn't surface in the sidebar or on the phone.","4. Expected: the run appears as an open/continuable session. Actual: it doesn't appear in either place."],"workaround":null,"fixed_in":null,"score":143.7,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openai/codex#48212","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":48212,"title":"[Linux Desktop] 26.924.20706 Codex tasks stuck on \"Starting your task\"; CLI works","url":"https://github.com/openai/codex/issues/48212","state":"closed","state_reason":"completed","labels":["bug","app","app-server"],"created":"2026-09-25T20:54:12Z","updated":"2026-09-28T21:12:51Z","closed":"2026-09-28T21:12:50Z","reactions":33,"thumbs_up":33,"comments":40,"severity":"medium","severity_reasons":["engagement 73","class:self-host/headless","class:crash/hang"],"topic":"self-host/headless","classes":["self-host/headless","crash/hang","install/update","auth/session"],"cosmetic_ui":false,"summary":"26.924.20706 Linux 6.6.141-09476-g954adab60416 x8664 unknown After updating ChatGPT Desktop on Linux to 26.924.20706, all local Codex tasks remain indefinitely on “Starting your task”. The standalone Codex CLI works correctly with the same…","versions":["26.924.20706","2.47.3","0.158.0-alpha.2."],"error_snippets":["thread_hydration outcome=failure reason=timeout ... wallDurationMs=120005","Worktree archive will be retried errorMessage=\"Git is unavailable\"","However, Git is available and works normally in the same Linux environment:","4. Submit a trivial read-only prompt, for example:","The desktop app should detect the installed Git binary at /usr/bin/git and should not report “Git is unavailable” when Git works normally from the same Linux environment."],"workaround":null,"fixed_in":null,"score":143.7,"first_seen":"2026-10-18T04:34:23Z"},{"id":"anthropics/claude-code#102142","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":102142,"title":"[BUG] Desktop 2.37337.1: Code tab Routines section hidden after the sidebar update, reset to hidden on every launch","url":"https://github.com/anthropics/claude-code/issues/102142","state":"open","state_reason":null,"labels":[],"created":"2026-10-14T18:53:04Z","updated":"2026-10-22T12:50:14Z","closed":null,"reactions":14,"thumbs_up":14,"comments":21,"severity":"medium","severity_reasons":["engagement 35","class:install/update"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"After updating the Claude desktop app, the Routines section is gone from the Code tab sidebar. My local routines still exist and still run on schedule; only the section is missing. The sidebar has no control to show it again. What I found…","versions":["2.1.296"],"error_snippets":["epitaxyPrefs[\"dframe-code-sections\"]","After updating the Claude desktop app, the Routines section is gone from the Code tab sidebar. My local routines still exist and still run on schedule; only the section is missing. The sidebar has no control to show it a"],"workaround":null,"fixed_in":null,"score":142.7,"first_seen":"2026-10-18T04:33:44Z"},{"id":"anomalyco/opencode#33742","harness":"opencode","harness_name":"OpenCode","repo":"anomalyco/opencode","number":33742,"title":"OpenCode v1.17.10 crashes with Bun segmentation fault on Windows; v1.17.9 appears stable","url":"https://github.com/anomalyco/opencode/issues/33742","state":"closed","state_reason":"not_planned","labels":[],"created":"2026-06-25T00:56:41Z","updated":"2026-10-11T17:22:29Z","closed":"2026-10-11T17:22:29Z","reactions":46,"thumbs_up":46,"comments":63,"severity":"high","severity_reasons":["title:breakage"],"topic":"crash/hang","classes":["crash/hang","install/update","network"],"cosmetic_ui":false,"summary":"After upgrading OpenCode to v1.17.10, OpenCode started crashing with a native Bun segmentation fault on Windows. Downgrading to v1.17.9 appears to stop the fatal crash under the same repository/configuration/session pattern. This looks…","versions":["1.17.10","1.17.9"],"error_snippets":["panic(main thread): Segmentation fault at address 0x2366F192AA2","Segmentation fault at address 0x7876BAA2","npm warn cleanup Failed to remove some directories","EPERM: operation not permitted, unlink ... opencode-windows-x64\\bin\\opencode.exe","ProviderModelNotFoundError"],"workaround":{"text":"Downgrading to `opencode-ai@1.17.9` appears to avoid the fatal crash under the same environment and workflow.","source":"issue body","url":"https://github.com/anomalyco/opencode/issues/33742"},"fixed_in":null,"score":141.8,"first_seen":"2026-10-19T18:38:33Z"},{"id":"anthropics/claude-code#93782","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":93782,"title":"[BUG] Regression in 2.1.269: dictation-tool paste (clipboard + simulated Ctrl+V) not inserted in VS Code integrated terminal (WSL2) — 2.1.268 works","url":"https://github.com/anthropics/claude-code/issues/93782","state":"open","state_reason":null,"labels":["bug","has repro","area:tui","area:ide","platform:vscode","regression","platform:wsl"],"created":"2026-09-12T07:50:57Z","updated":"2026-10-07T05:06:52Z","closed":null,"reactions":9,"thumbs_up":9,"comments":23,"severity":"high","severity_reasons":["label:regression","title:breakage"],"topic":"self-host/headless","classes":["self-host/headless","install/update","crash/hang"],"cosmetic_ui":false,"summary":"After auto-updating to 2.1.269, text inserted by a voice dictation tool (Wispr Flow — clipboard + simulated Ctrl+V) is no longer inserted into the Claude Code prompt when Claude Code runs in the VS Code integrated terminal (Remote-WSL)…","versions":["2.1.269","2.1.268","1.6.827"],"error_snippets":["editor.accessibilitySupport","voice.enabled: false","Running **2.1.268 in the exact same terminal works**, so this is a regression introduced in 2.1.269.","So the failure is exactly the intersection *2.1.269 × VS Code integrated terminal*; neither factor alone reproduces it.","The 2.1.269 changelog includes terminal keyboard-input fixes (F1/F2/F4 in kitty-protocol terminals, Delete in st, Alt+arrows in rxvt-unicode, Shift+punctuation in WezTerm). The regression looks like a side effect of thos"],"workaround":{"text":"Pin 2.1.268 (`ln -sfn ~/.local/share/claude/versions/2.1.268 ~/.local/bin/claude` + `DISABLE_AUTOUPDATER=1`), or run Claude Code in Windows Terminal instead of the VS Code integrated terminal.","source":"issue body","url":"https://github.com/anthropics/claude-code/issues/93782"},"fixed_in":null,"score":141.2,"first_seen":"2026-10-18T04:33:44Z"},{"id":"github/copilot-cli#1305","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":1305,"title":"Support CIMD for Remote OAuth MCP Servers","url":"https://github.com/github/copilot-cli/issues/1305","state":"closed","state_reason":"completed","labels":["area:authentication","area:mcp"],"created":"2026-02-05T06:48:17Z","updated":"2026-10-19T20:11:29Z","closed":"2026-09-27T10:13:07Z","reactions":39,"thumbs_up":39,"comments":7,"severity":"medium","severity_reasons":["engagement 46","class:self-host/headless","class:auth/session"],"topic":"self-host/headless","classes":["self-host/headless","auth/session","mcp/tools"],"cosmetic_ui":false,"summary":"In 0.0.389 we added support for Remote MCP Servers protected by OAuth that support the DCR (Dynamic Client Registration) standard. DCR allows an OAuth client to register itself with an OAuth server \"just in time\" instead of requiring the…","versions":["0.0.389"],"error_snippets":["I think you, @ellismg, asked to let you know which MCP servers people are trying to use that may need this CIMD feature.","In addition to Atlassian (already mentioned), my team also uses expo, which also requires CIMD for its MCP server.","https://docs.expo.dev/mcp/"],"workaround":null,"fixed_in":null,"score":140.1,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openai/codex#48076","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":48076,"title":"blank windows terminal window","url":"https://github.com/openai/codex/issues/48076","state":"open","state_reason":null,"labels":["bug","windows-os","app","app-server"],"created":"2026-09-25T12:01:03Z","updated":"2026-10-07T06:17:55Z","closed":null,"reactions":23,"thumbs_up":23,"comments":16,"severity":"medium","severity_reasons":["engagement 39","class:self-host/headless"],"topic":"self-host/headless","classes":["self-host/headless","install/update"],"cosmetic_ui":false,"summary":"After updating Codex on September 25, 2026, a blank Windows Terminal window for codex-code-mode-host.exe repeatedly opens and closes while I use Codex. This did not happen before the update. The window title points to the app-server-daemon…","versions":["0.157.0"],"error_snippets":[],"workaround":null,"fixed_in":null,"score":139.8,"first_seen":"2026-10-19T18:38:33Z"},{"id":"anomalyco/opencode#11112","harness":"opencode","harness_name":"OpenCode","repo":"anomalyco/opencode","number":11112,"title":"always stuck at  “Preparing write...”","url":"https://github.com/anomalyco/opencode/issues/11112","state":"closed","state_reason":"duplicate","labels":[],"created":"2026-01-29T09:05:19Z","updated":"2026-10-01T09:35:42Z","closed":"2026-10-01T09:35:42Z","reactions":52,"thumbs_up":52,"comments":83,"severity":"medium","severity_reasons":["engagement 135","class:crash/hang"],"topic":"crash/hang","classes":["crash/hang"],"cosmetic_ui":false,"summary":"My Prometheus always stuck at “Preparing write...”。","versions":[],"error_snippets":["Tool execution aborted"],"workaround":null,"fixed_in":null,"score":135.7,"first_seen":"2026-10-19T18:38:33Z"},{"id":"anthropics/claude-code#99217","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":99217,"title":"Mod ui.render (AbovePrompt/Pane) not drawn in desktop/iOS views of a Remote Control session","url":"https://github.com/anthropics/claude-code/issues/99217","state":"open","state_reason":null,"labels":["bug","area:plugins"],"created":"2026-10-03T12:20:51Z","updated":"2026-10-22T12:45:41Z","closed":null,"reactions":11,"thumbs_up":11,"comments":10,"severity":"medium","severity_reasons":["engagement 21","class:self-host/headless","class:mcp/tools"],"topic":"self-host/headless","classes":["self-host/headless","mcp/tools"],"cosmetic_ui":false,"summary":"A mod's ui.render output (e.g. an AbovePrompt band with a Button) is drawn in the terminal but never appears in the Claude desktop app or the Claude iOS app when they view the same session through Remote Control. Only mod-initiated…","versions":["2.1.287"],"error_snippets":["The docs list `AbovePrompt` and `Pane` as rendered on \"Terminal, Desktop\" (mods reference, render sites) and say both \"work in\" the terminal and the Desktop app. The API also lists `desktop` and `mobile` surfaces. I expe"],"workaround":null,"fixed_in":null,"score":133.8,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openai/codex#48059","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":48059,"title":"[Windows][Codex CLI 0.157.0] Terminal windows repeatedly pop up during normal use","url":"https://github.com/openai/codex/issues/48059","state":"closed","state_reason":"duplicate","labels":["bug","windows-os","CLI","app-server"],"created":"2026-09-25T10:36:00Z","updated":"2026-10-15T15:35:13Z","closed":"2026-09-29T06:55:41Z","reactions":53,"thumbs_up":53,"comments":27,"severity":"medium","severity_reasons":["engagement 80","class:self-host/headless"],"topic":"self-host/headless","classes":["self-host/headless","always-on/daemon","install/update","mcp/tools"],"cosmetic_ui":false,"summary":"Codex CLI: codex-cli 0.157.0 (global npm installation) Managed app-server daemon: 0.157.0-x8664-pc-windows-msvc Codex Desktop 26.917.9434.0 is also installed. The process chain captured below belongs to the active CLI session; I have not…","versions":["0.157.0"],"error_snippets":["The exact executable shown in every popup has not yet been captured, so this process chain is supporting evidence rather than proof that every popup has the same source.","A similar report about daemon-side console spawning is #44768. This issue records the symptom on CLI 0.157.0 and the current local environment. No personal paths, account details, or raw logs are included."],"workaround":null,"fixed_in":null,"score":131.4,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openclaw/openclaw#76493","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":76493,"title":"Allow SecretRef objects in `mcp.servers[].env` schema (currently `string|number|boolean` only)","url":"https://github.com/openclaw/openclaw/issues/76493","state":"open","state_reason":null,"labels":["P2","clawsweeper:no-new-fix-pr","clawsweeper:needs-maintainer-review","clawsweeper:needs-product-decision","clawsweeper:needs-security-review","clawsweeper:source-repro","impact:security","impact:auth-provider","issue-rating: 🦞 diamond lobster"],"created":"2026-05-03T06:10:46Z","updated":"2026-10-20T16:46:58Z","closed":null,"reactions":5,"thumbs_up":4,"comments":12,"severity":"critical","severity_reasons":["label:security"],"topic":"self-host/headless","classes":["self-host/headless","auth/session","mcp/tools","always-on/daemon","install/update"],"cosmetic_ui":false,"summary":"The schema for mcp.servers[].env (and mcp.servers[].headers) is: This means SecretRef objects are rejected — operators following security best practices around secret indirection (openclaw config set --ref-source env --ref-id ...) hit a…","versions":["2026.4.29"],"error_snippets":["The 3 most common MCP integrations (notion, github, postgres) each require at least one secret in `mcp.servers[].env`. With the current schema, these secrets MUST be plaintext in `openclaw.json`.","This is the architectural floor for \"move secrets out of openclaw.json\" work. Our audit identified 7 plaintext secrets in `mcp.servers[].env` that cannot be moved to env-var indirection because the schema rejects SecretR"],"workaround":{"text":"The only documented workaround is the global `$OPENCLAW_STATE_DIR/.env`.","source":"comment by @davidcittadini","url":"https://github.com/openclaw/openclaw/issues/76493#issuecomment-6251467513"},"fixed_in":null,"score":131.3,"first_seen":"2026-10-19T18:38:33Z"},{"id":"github/copilot-cli#970","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":970,"title":"Copilot app blocked by macOS Gatekeeper under corporate security policy","url":"https://github.com/github/copilot-cli/issues/970","state":"open","state_reason":null,"labels":["area:installation"],"created":"2026-01-14T22:24:40Z","updated":"2026-10-12T15:35:04Z","closed":null,"reactions":23,"thumbs_up":23,"comments":8,"severity":"medium","severity_reasons":["engagement 31","class:install/update"],"topic":"install/update","classes":["install/update","network"],"cosmetic_ui":false,"summary":"Everytime after upgrading copilot cli through HomeBrew the following message appears: Apple could not verify “copilot” is free of malware that may harm your Mac or compromise your privacy. Afterwards, to get it working correctly, I need to…","versions":["0.0.381"],"error_snippets":["Everytime after upgrading copilot cli through HomeBrew the following message appears:","Afterwards, to get it working correctly, I need to go to Privacy & Security to explicitly allow \"copilot\"","1. On a corporate‑managed Mac with Gatekeeper enforced to App Store & Known Developers by profile."],"workaround":null,"fixed_in":null,"score":130.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"continuedev/continue#3753","harness":"continue","harness_name":"Continue","repo":"continuedev/continue","number":3753,"title":"Intermittent Copy/Paste Not Working and Extension Host Crashes in VSCode [specifically for `.sh` files]","url":"https://github.com/continuedev/continue/issues/3753","state":"closed","state_reason":"not_planned","labels":["area:autocomplete","kind:bug","ide:vscode","priority:high","stale","os:linux","released"],"created":"2025-01-17T15:35:01Z","updated":"2026-10-19T01:50:52Z","closed":"2026-10-19T01:50:52Z","reactions":9,"thumbs_up":8,"comments":79,"severity":"high","severity_reasons":["label:high","title:breakage"],"topic":"crash/hang","classes":["crash/hang","mcp/tools","install/update"],"cosmetic_ui":false,"summary":"I believe this is a bug. I'll try to join the Continue Discord for questions I'm not able to find an open issue that reports the same bug I've seen the troubleshooting guide on the Continue Docs After using the Continue VSCode extension…","versions":["0.8.66","565.77","12.6.85","0.8.65"],"error_snippets":["- NVIDIA Drivers: 565.77","- Other Enabled Extensions: Bash Debug, Bash IDE, ShellCheck","\"description\": \"Export the current chat session to markdown\"","[info] Extension host (LocalProcess pid: 336281) is unresponsive.","[info] UNRESPONSIVE extension host: starting to profile NOW"],"workaround":null,"fixed_in":null,"score":129.8,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openai/codex#48235","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":48235,"title":"401 Unauthorized across Codex Desktop and CLI after working normally all day","url":"https://github.com/openai/codex/issues/48235","state":"closed","state_reason":"duplicate","labels":["bug","auth","app"],"created":"2026-09-25T22:51:18Z","updated":"2026-09-25T23:34:49Z","closed":"2026-09-25T23:07:10Z","reactions":56,"thumbs_up":56,"comments":36,"severity":"medium","severity_reasons":["engagement 92","class:auth/session"],"topic":"auth/session","classes":["auth/session","self-host/headless"],"cosmetic_ui":false,"summary":"Codex Desktop and Codex CLI both began failing suddenly with 401 Unauthorized after working normally all day. Re-authenticating in both clients does not resolve the issue. Codex CLI: 0.155.1 Login status: Logged in using ChatGPT Remote…","versions":["0.155.1"],"error_snippets":["unexpected status 401 Unauthorized: Incorrect API key provided: sk-svcac[REDACTED].","Codex Desktop and Codex CLI both began failing suddenly with 401 Unauthorized after working normally all day. Re-authenticating in both clients does not resolve the issue.","The error included a Cloudflare ray ID and request ID, but the credential itself is intentionally redacted here.","5. Same 401 persists in both environments","Please investigate whether there is an auth-routing/session issue affecting ChatGPT-authenticated Codex traffic even though the public status page currently shows green."],"workaround":null,"fixed_in":null,"score":129.6,"first_seen":"2026-10-19T18:38:33Z"},{"id":"github/copilot-cli#3100","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":3100,"title":"HTTP MCP server with Bearer token fails OAuth discovery instead of falling back to headers auth","url":"https://github.com/github/copilot-cli/issues/3100","state":"open","state_reason":null,"labels":["area:authentication","area:mcp"],"created":"2026-05-04T07:20:56Z","updated":"2026-10-16T15:21:31Z","closed":null,"reactions":11,"thumbs_up":11,"comments":2,"severity":"medium","severity_reasons":["class:self-host/headless","class:auth/session"],"topic":"self-host/headless","classes":["self-host/headless","auth/session","mcp/tools","network"],"cosmetic_ui":false,"summary":"When an HTTP MCP server is configured in .mcp.json with \"type\": \"http\" and \"headers\": { \"Authorization\": \"Bearer \" }, the CLI attempts OAuth discovery (/.well-known/oauth-authorization-server ) and fails with: MCPOAuthError: Failed to…","versions":[],"error_snippets":["MCPOAuthError: Failed to discover authorization server metadata","The CLI does not fall back to using the provided Bearer token from headers, causing a hard auth failure even though valid credentials are present in the config.","4. Pay attention to the error","Expected behavior: If OAuth discovery fails and headers contains Authorization: Bearer, the CLI should skip OAuth and proceed to MCP initialize using the provided token.","Workaround: Manually enabling the server via /mcp enable bypasses OAuth discovery and connects successfully using the Bearer token."],"workaround":{"text":"Manually enabling the server via /mcp enable bypasses OAuth discovery and connects successfully using the Bearer token.","source":"issue body","url":"https://github.com/github/copilot-cli/issues/3100"},"fixed_in":null,"score":129.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"NousResearch/hermes-agent#34352","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":34352,"title":"Solving the Multi-Tenant Hermes Problem","url":"https://github.com/NousResearch/hermes-agent/issues/34352","state":"open","state_reason":null,"labels":["type/feature","comp/agent","comp/gateway","tool/memory","P3","needs-decision","sweeper:risk-session-state"],"created":"2026-05-29T04:13:02Z","updated":"2026-10-17T09:48:22Z","closed":null,"reactions":17,"thumbs_up":7,"comments":25,"severity":"medium","severity_reasons":["engagement 42","class:always-on/daemon"],"topic":"always-on/daemon","classes":["always-on/daemon","network"],"cosmetic_ui":false,"summary":"Multiplayer agentic AI is the future. Hermes can and should lead. TL;DR: Memory operations bypass the hook system entirely, making tenant isolation impossible without forking core — we've been running a fix in production for months with…","versions":[],"error_snippets":["# Called before every memory read and write operation.","flag missing from SSE**: signal-cli's SSE endpoint doesn't include","_TL;DR: Memory operations bypass the hook system entirely, making tenant isolation impossible without forking core — we've been running a fix in production for months with numerous multi-tenant agents in different contex","Every Hermes deployment beyond personal use hits the same wall: one agent = one tenant. Memory is global, sessions don't scope by tenant, and there is no isolation between groups, channels, or users. Run a Hermes agent i","The Hermes's plugin/hook architecture is genuinely strong. The 17-hook system — particularly `pre_gateway_dispatch` for routing and gating — handles ~60% of what a multi-tenant deployment needs. Per-adapter environment v"],"workaround":null,"fixed_in":null,"score":128.5,"first_seen":"2026-10-18T04:29:32Z"},{"id":"openclaw/openclaw#159596","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":159596,"title":"[Bug]: Gateway memory sawtooth on 2026.9.6 — prepared-model-catalog worker grows to the full heap ceiling; ~200 critical memory-pressure events/day","url":"https://github.com/openclaw/openclaw/issues/159596","state":"open","state_reason":null,"labels":["stale","clawsweeper:no-new-fix-pr","clawsweeper:needs-maintainer-review","impact:crash-loop","P0","issue-rating: 🦪 silver shellfish"],"created":"2026-09-27T10:40:22Z","updated":"2026-10-25T12:08:27Z","closed":null,"reactions":4,"thumbs_up":3,"comments":16,"severity":"critical","severity_reasons":["label:p0"],"topic":"always-on/daemon","classes":["always-on/daemon","crash/hang","self-host/headless","install/update","mcp/tools"],"cosmetic_ui":false,"summary":"Regression / performance On OpenClaw 2026.9.6 the Gateway exhibits a repeating memory \"sawtooth\": process RSS climbs steadily for several minutes, trips a memory pressure: level=critical event, an idle worker is retired under pressure, RSS…","versions":["2026.9.6","16.6","6.07","2.8"],"error_snippets":["openclaw gateway diagnostics export","On OpenClaw 2026.9.6 the Gateway exhibits a repeating memory \"sawtooth\": process RSS","climbs steadily for several minutes, trips a `memory pressure: level=critical` event,","Observed ~200+ critical pressure events in a single day — all `level=critical`, none","`warning` — with concurrent event-loop stalls."],"workaround":null,"fixed_in":null,"score":128.4,"first_seen":"2026-10-18T04:33:05Z"},{"id":"cline/cline#8636","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":8636,"title":"Thinking section not displayed in v3.50.0","url":"https://github.com/cline/cline/issues/8636","state":"open","state_reason":null,"labels":["VS Code","Bot Responded","Regression"],"created":"2026-01-15T10:36:06Z","updated":"2026-10-10T22:35:11Z","closed":null,"reactions":17,"thumbs_up":3,"comments":35,"severity":"high","severity_reasons":["label:regression"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"VSCode Extension The \"Thinking\" section (extended thinking / chain-of-thought display) is no longer visible in Cline v3.50.0. This feature was working correctly in v3.45.1. Install Cline v3.50.0 Start a new conversation with Claude Ask any…","versions":["3.50.0"],"error_snippets":["VSCode Extension","The \"Thinking\" section (extended thinking / chain-of-thought display) is no longer visible in Cline v3.50.0. This feature was working correctly in v3.45.1."],"workaround":null,"fixed_in":null,"score":127.7,"first_seen":"2026-10-19T18:38:33Z"},{"id":"NousResearch/hermes-agent#142017","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":142017,"title":"gateway /model pick on a profile-following session crashes (AttributeError: 'tuple' object has no attribute 'get') when the session carries config secrets","url":"https://github.com/NousResearch/hermes-agent/issues/142017","state":"open","state_reason":null,"labels":["type/bug","comp/gateway","area/auth","P2","area/sessions","area/profiles"],"created":"2026-10-20T11:11:26Z","updated":"2026-10-24T02:49:38Z","closed":null,"reactions":1,"thumbs_up":1,"comments":9,"severity":"high","severity_reasons":["title:breakage"],"topic":"always-on/daemon","classes":["always-on/daemon","crash/hang","auth/session","install/update","network"],"cosmetic_ui":false,"summary":"A /model pick made in any gateway chat (Telegram, Bot Chat, TUI) on a session that follows the profile config fails with an internal error (\"Model switch failed\") whenever the session policy carries private config secrets…","versions":[],"error_snippets":["AttributeError: 'tuple' object has no attribute 'get'","provider_secrets, retired, moved = ({}, (), {}) if old.follows_profile else _freeze_selected_provider(...)","secrets = {moved.get(path[:2], path[:2]) + path[2:]: ...} # crashes when secrets is non-empty","— which the earlier line initialized as a **dict** (the secrets path remap from","A `/model` pick made in any gateway chat (Telegram, Bot Chat, TUI) on a session that follows the profile config fails with an internal error (\"Model switch failed\") whenever the session policy carries private config secr"],"workaround":null,"fixed_in":null,"score":126.4,"first_seen":"2026-10-25T09:11:37Z"},{"id":"NousResearch/hermes-agent#50148","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":50148,"title":"Add Nextcloud Talk as a messaging platform adapter (working implementation, ready to PR)","url":"https://github.com/NousResearch/hermes-agent/issues/50148","state":"open","state_reason":null,"labels":["type/feature","comp/gateway","P3"],"created":"2026-06-21T12:49:14Z","updated":"2026-10-17T00:19:02Z","closed":null,"reactions":20,"thumbs_up":16,"comments":7,"severity":"medium","severity_reasons":["engagement 27","class:always-on/daemon"],"topic":"always-on/daemon","classes":["always-on/daemon","self-host/headless","install/update","crash/hang","auth/session","network"],"cosmetic_ui":false,"summary":"I'd like to contribute a Nextcloud Talk messaging-platform adapter for the gateway — the same shape as the existing Mattermost / Signal / Telegram adapters: Hermes both receives messages from and sends replies to a Talk conversation, fully…","versions":[],"error_snippets":["the same shape as the existing Mattermost / Signal / Telegram adapters: Hermes both","messaging channel with **no third-party service in the loop** — everything stays on their","own infrastructure. It slots into the existing multi-platform gateway exactly like the","so it follows existing conventions.","Talk webhooks and posts replies in a real room."],"workaround":null,"fixed_in":null,"score":126.4,"first_seen":"2026-10-18T04:29:32Z"},{"id":"openclaw/openclaw#153257","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":153257,"title":"[Bug]: OpenClaw 2026.9.5 Turned a Stable Environment Into an 8-Hour Failure Recovery Session","url":"https://github.com/openclaw/openclaw/issues/153257","state":"open","state_reason":null,"labels":["bug","bug:crash","P1","impact:crash-loop","issue-rating: 🦪 silver shellfish","impact:ux-friction","clawsweeper:manual-only"],"created":"2026-09-19T23:36:16Z","updated":"2026-10-16T03:42:44Z","closed":null,"reactions":2,"thumbs_up":1,"comments":41,"severity":"critical","severity_reasons":["label:crash-loop"],"topic":"crash/hang","classes":["crash/hang","always-on/daemon","install/update"],"cosmetic_ui":false,"summary":"Crash (process/app exits or hangs) I genuinely regret upgrading to OpenClaw 2026.9.5. Before this update, my environment was stable. After installing 9.5, I spent approximately 8 continuous hours investigating problems that simply did not…","versions":["2026.9.5","9.5","148.72","2.177"],"error_snippets":["SQLite transaction admission was refused","errorCode=UNAVAILABLE","Crash (process/app exits or hangs)","Before this update, my environment was stable. After installing 9.5, I spent approximately **8 continuous hours** investigating problems that simply did not exist before the upgrade.","I started questioning my configuration, my infrastructure, my providers, my containers, my model setup, and eventually began changing settings in an environment that had already been working correctly."],"workaround":null,"fixed_in":null,"score":126.4,"first_seen":"2026-10-18T04:33:05Z"},{"id":"openai/codex#49988","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":49988,"title":"Code extension intermittently drops submitted messages after update.","url":"https://github.com/openai/codex/issues/49988","state":"closed","state_reason":"completed","labels":["bug","extension"],"created":"2026-10-01T14:02:43Z","updated":"2026-10-10T02:12:43Z","closed":"2026-10-04T02:57:08Z","reactions":50,"thumbs_up":49,"comments":53,"severity":"medium","severity_reasons":["engagement 103","class:install/update"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"After updating the Codex VS Code extension on October 1, 2026, pressing Enter frequently clears the composer, but the message does not appear in the conversation and Codex does not respond. This happens even when Codex is idle. Repeating…","versions":["26.928.31416","0.159.2","26.917.62051"],"error_snippets":["After updating the Codex VS Code extension on October 1, 2026,","pressing Enter frequently clears the composer, but the message","This happens even when Codex is idle. Repeating the same message","several times eventually succeeds.","Downgrading to extension version 26.917.62051 stops the problem."],"workaround":{"text":"Downgrading to extension version 26.917.62051 stops the problem.","source":"issue body","url":"https://github.com/openai/codex/issues/49988"},"fixed_in":null,"score":126.0,"first_seen":"2026-10-18T04:31:06Z"},{"id":"kirodotdev/Kiro#12054","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":12054,"title":"Remote MCP (Atlassian, OAuth) times out after \"Authorization Successful\"","url":"https://github.com/kirodotdev/Kiro/issues/12054","state":"open","state_reason":null,"labels":["mcp","auth","pending-maintainer-response"],"created":"2026-10-21T14:18:23Z","updated":"2026-10-22T14:23:50Z","closed":null,"reactions":1,"thumbs_up":1,"comments":9,"severity":"medium","severity_reasons":["class:self-host/headless","class:auth/session"],"topic":"self-host/headless","classes":["self-host/headless","auth/session","mcp/tools","install/update","network"],"cosmetic_ui":false,"summary":"Windows 11 (NT 10.0.26300) Kiro: 1.4.13 · Windows 11 (NT 10.0.26300) Server: https://mcp.atlassian.com/v2/mcp (remote, OAuth via DCR, no custom config) What happens Click Retry → browser opens Atlassian consent → \"Authorization Successful\"…","versions":["10.0.26300","1.4.13","2.4"],"error_snippets":["~50 s, then \"MCP server 'atlassian' connection timed out after 60000ms\".","MCP Servers panel shows \"Connection Failed\".","fails with \"Failed to verify code verifier\".","Example (2026-10-21, UTC):","14:06:01 Failed to connect: connection timed out after 60000ms"],"workaround":null,"fixed_in":null,"score":125.4,"first_seen":"2026-10-22T09:27:06Z"},{"id":"anomalyco/opencode#50650","harness":"opencode","harness_name":"OpenCode","repo":"anomalyco/opencode","number":50650,"title":"desktop: custom provider save always throws \"unavailable on this server\"","url":"https://github.com/anomalyco/opencode/issues/50650","state":"open","state_reason":null,"labels":[],"created":"2026-09-22T14:17:30Z","updated":"2026-10-23T21:12:52Z","closed":null,"reactions":15,"thumbs_up":13,"comments":13,"severity":"medium","severity_reasons":["engagement 28","class:self-host/headless"],"topic":"self-host/headless","classes":["self-host/headless","install/update"],"cosmetic_ui":false,"summary":"OpenCode Desktop exposes the \"Custom OpenAI-compatible provider\" form in Settings → Providers, but the save handler unconditionally throws provider.custom.unavailable, so the flow can never succeed on any server — including the app's own…","versions":["2.0.14"],"error_snippets":["throw Error(a.t(`provider.custom.unavailable`))","{\"id\":\"local\",\"name\":\"My Local Server\",\"activation\":\"enabled\", ...}","OpenCode Desktop exposes the \"Custom OpenAI-compatible provider\" form in Settings → Providers, but the save handler unconditionally throws `provider.custom.unavailable`, so the flow can never succeed on any server — incl","3. Select the custom provider entry (\"Custom OpenAI-compatible provider\").","It throws for every server, so the dialog is reachable but unusable."],"workaround":null,"fixed_in":null,"score":124.6,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openai/codex#48208","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":48208,"title":"[Linux Desktop][Regression] Codex UI hangs after update; thread_hydration times out while app-server remains responsive","url":"https://github.com/openai/codex/issues/48208","state":"closed","state_reason":"completed","labels":["bug","app","app-server","Linux"],"created":"2026-09-25T20:36:08Z","updated":"2026-09-29T20:05:18Z","closed":"2026-09-28T21:17:36Z","reactions":17,"thumbs_up":17,"comments":28,"severity":"high","severity_reasons":["title:breakage"],"topic":"self-host/headless","classes":["self-host/headless","install/update","crash/hang","network"],"cosmetic_ui":false,"summary":"After the most recent Codex/ChatGPT desktop update on Ubuntu, the Codex desktop UI opens but a Codex chat remains stuck on the loading spinner and cannot be interacted with. This started immediately after the update. Bundled Codex CLI…","versions":["0.158.0-alpha.2"],"error_snippets":["name=thread_hydration outcome=failure reason=timeout durationMs=304709.7 wallDurationMs=304710","name=thread_hydration outcome=failure reason=timeout durationMs=314626.2 wallDurationMs=314627","Worktree archive will be retried errorMessage=\"Git is unavailable\"","After the most recent Codex/ChatGPT desktop update on Ubuntu, the Codex desktop UI opens but a Codex chat remains stuck on the loading spinner and cannot be interacted with.","2. Open Codex / an existing Codex chat."],"workaround":null,"fixed_in":null,"score":124.4,"first_seen":"2026-10-18T04:34:23Z"},{"id":"omacom/omarchy#6754","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":6754,"title":"Waybar update indicator stuck: pre-release tag (v4.0.0-beta3) always reported as an update even when up to date","url":"https://github.com/omacom/omarchy/issues/6754","state":"open","state_reason":null,"labels":["bug"],"created":"2026-08-12T14:50:47Z","updated":"2026-10-08T03:56:19Z","closed":null,"reactions":12,"thumbs_up":12,"comments":7,"severity":"medium","severity_reasons":["engagement 19","class:install/update","class:crash/hang"],"topic":"install/update","classes":["install/update","crash/hang","self-host/headless"],"cosmetic_ui":false,"summary":"Omarchy: 3.8.3 | Arch Linux | Waybar custom/update module: \"exec\": \"omarchy-update-available\" On Omarchy, Waybar shows the \"updates available\" icon when the system is not up to date. This icon is controlled by bin/omarchy-update-available…","versions":["4.0.0-beta3","3.8.3","3.8.4"],"error_snippets":["Omarchy: 3.8.3 | Arch Linux | Waybar custom/update module: \"exec\": \"omarchy-update-available\"","The problem is that the icon **never disappears**, even when the system is fully up to date.","If the remote version differs from the local version, the script prints `update available` and returns exit code `0`, causing Waybar to display the update icon.","This creates an endless loop:","In other words, the update checker is reporting an update that the stable channel **cannot and will not install**."],"workaround":null,"fixed_in":null,"score":122.7,"first_seen":"2026-10-19T18:38:33Z"},{"id":"NousResearch/hermes-agent#134220","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":134220,"title":"Bundled plugin solstice fails to load on fresh installs: No module named 'httpx' (startup warnings paint over TUI)","url":"https://github.com/NousResearch/hermes-agent/issues/134220","state":"open","state_reason":null,"labels":["type/bug","comp/cli","comp/tui","comp/plugins","P3","sweeper:risk-compatibility","area/install-update"],"created":"2026-10-06T22:22:08Z","updated":"2026-10-08T16:34:22Z","closed":null,"reactions":8,"thumbs_up":8,"comments":7,"severity":"medium","severity_reasons":["engagement 15","class:install/update","class:mcp/tools"],"topic":"install/update","classes":["install/update","mcp/tools","auth/session","network"],"cosmetic_ui":false,"summary":"On startup, this warning is printed (twice in my case) before the agent begins processing: The messages render over already-drawn TUI content, garbling the screen. After the warnings, processing starts and works normally — solstice is a…","versions":["3.14","0.28.1","0.1.0"],"error_snippets":["Failed to load bundled provider plugin solstice: No module named 'httpx'","raise ModuleNotFoundError(\"No module named 'httpx'\", name=name)","import solstice # -> ModuleNotFoundError: No module named 'httpx'","resolution at plugin load time. Repro on an environment without","hermes_cli/plugins_loader.py:537"],"workaround":null,"fixed_in":null,"score":122.6,"first_seen":"2026-10-18T04:29:32Z"},{"id":"kirodotdev/Kiro#1533","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":1533,"title":"Unable to Authenticate (Sign-In) - Browser Not Launching on Linux Mint (Debian/Ubuntu .deb install)","url":"https://github.com/kirodotdev/Kiro/issues/1533","state":"closed","state_reason":"completed","labels":["auth"],"created":"2025-07-29T05:28:07Z","updated":"2026-09-24T13:05:54Z","closed":"2026-09-24T13:05:54Z","reactions":9,"thumbs_up":7,"comments":60,"severity":"medium","severity_reasons":["engagement 69","class:install/update","class:auth/session"],"topic":"install/update","classes":["install/update","auth/session","crash/hang","network"],"cosmetic_ui":false,"summary":"Linux Mint 22.1 Cinnamon When attempting to sign in to Kiro IDE (via Google, GitHub, or any other authentication provider), the IDE displays a message indicating it will open a browser for sign-in, but no browser window actually launches…","versions":["22.1","0.1.25"],"error_snippets":["5. Expected: A new browser tab or window opens to the authentication page.","1. I expected it to launch a new browser tab and bring me to the authentication page."],"workaround":null,"fixed_in":null,"score":122.3,"first_seen":"2026-10-19T18:38:33Z"},{"id":"NousResearch/hermes-agent#110374","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":110374,"title":"Slack: status indicator silently broken with slack-sdk >= 3.44 (agents.sessions.setStatus rejects free-text and empty status)","url":"https://github.com/NousResearch/hermes-agent/issues/110374","state":"open","state_reason":null,"labels":["type/bug","comp/plugins","platform/slack","P3"],"created":"2026-09-13T23:15:40Z","updated":"2026-10-16T10:06:37Z","closed":null,"reactions":16,"thumbs_up":16,"comments":8,"severity":"high","severity_reasons":["title:breakage"],"topic":"mcp/tools","classes":["mcp/tools","install/update","crash/hang"],"cosmetic_ui":false,"summary":"After updating past slack-sdk 3.44.0, the Slack working-status indicator (\"is thinking…\", live per-tool status via livestatus / typingstatustext) silently disappears on every turn. All status calls fail with invalidarguments; status…","versions":["3.44","3.44.0","3.44.1","1.30.0"],"error_snippets":["must be a valid enum value [json-pointer:/status]","does NOT clear when the app replies (the legacy method auto-clears through the compat bridge). A turn that ends without explicitly setting","After updating past slack-sdk 3.44.0, the Slack working-status indicator (\"is thinking…\", live per-tool status via `live_status` / `typing_status_text`) silently disappears on every turn. All status calls fail with `inva","All of these are rejected (`must be a valid enum value [json-pointer:/status]`). The chooser changed the wire protocol without changing the payloads, and both affected paths — set (`_set_thread_status`) and clear (the em","Any deployment with slack-sdk >=3.44.0 gets no indicator and no logged error."],"workaround":null,"fixed_in":null,"score":120.1,"first_seen":"2026-10-18T04:29:32Z"},{"id":"openclaw/openclaw#179481","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":179481,"title":"[Bug]: doctor --fix refuses `transcript-directives` while the gateway is running and blocks every later migration step — no online repair path","url":"https://github.com/openclaw/openclaw/issues/179481","state":"open","state_reason":null,"labels":["P2","clawsweeper:source-repro","issue-rating: 🦞 diamond lobster","maturity:stable","impact:ux-friction"],"created":"2026-10-21T12:41:05Z","updated":"2026-10-22T09:02:37Z","closed":null,"reactions":2,"thumbs_up":1,"comments":11,"severity":"medium","severity_reasons":["class:always-on/daemon","class:install/update"],"topic":"always-on/daemon","classes":["always-on/daemon","install/update","auth/session","mcp/tools"],"cosmetic_ui":false,"summary":"After a recovery that moved the old session store aside, openclaw doctor stops at: The only holder of the agent DB is the running gateway itself. Deterministic on our host: Run OpenClaw 2026.10.5 as a systemd user service (gateway…","versions":["2026.10.5"],"error_snippets":["Failing check transcript-directives (step-refused): Skipped historical transcript directive migration:","OpenClawAgentDatabaseLeaseActiveError: Agent main database is still open in another process;","OpenClawAgentDatabaseLeaseActiveError","returns \"Automation not found\" for job ids that","3. Make the error say that the lease holder is the gateway and that stopping it is required."],"workaround":{"text":"`P2`: A large legacy-history installation cannot reliably complete offline repair at default limits, with an advanced-operator workaround and a focused repair already proposed. - `impact:ux-friction`: Repeated offline repair failures require downtime and manual runtime tuning before the operator…","source":"comment by @clawsweeper[bot]","url":"https://github.com/openclaw/openclaw/issues/179481#issuecomment-6321165328"},"fixed_in":null,"score":120.1,"first_seen":"2026-10-22T09:27:06Z"},{"id":"openai/codex#46110","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":46110,"title":"Linux sandbox rejects valid nsfs mount roots from snapd: `mountinfo path is not absolute`","url":"https://github.com/openai/codex/issues/46110","state":"open","state_reason":null,"labels":["bug","sandbox","CLI"],"created":"2026-09-17T04:31:38Z","updated":"2026-09-30T09:46:35Z","closed":null,"reactions":21,"thumbs_up":21,"comments":24,"severity":"medium","severity_reasons":["engagement 45"],"topic":"always-on/daemon","classes":["always-on/daemon","install/update","network"],"cosmetic_ui":false,"summary":"Codex 0.155.0-alpha.15 fails before executing any filesystem-restricted command on a native Ubuntu host when /proc/self/mountinfo contains valid nsfs entries created by snapd. The failure is: The affected mountinfo entries have an nsfs…","versions":["0.155.0-alpha.15","0.11.1"],"error_snippets":["error building bubblewrap command: mountinfo path is not absolute","status: failed","exit code: 1","output: error building bubblewrap command: mountinfo path is not absolute","return Err(io::Error::other(\"mountinfo path is not absolute\"));"],"workaround":null,"fixed_in":null,"score":119.4,"first_seen":"2026-10-18T04:34:23Z"},{"id":"NousResearch/hermes-agent#1855","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":1855,"title":"feat(terminal): Multi-backend terminal — local + N named remotes with persistent shell","url":"https://github.com/NousResearch/hermes-agent/issues/1855","state":"open","state_reason":null,"labels":["type/feature","backend/ssh","comp/tools","tool/terminal","P3","needs-decision"],"created":"2026-03-18T01:26:28Z","updated":"2026-09-23T18:55:08Z","closed":null,"reactions":17,"thumbs_up":17,"comments":11,"severity":"medium","severity_reasons":["engagement 28","class:self-host/headless"],"topic":"self-host/headless","classes":["self-host/headless","network"],"cosmetic_ui":false,"summary":"Today, Hermes Agent supports exactly one terminal backend at a time — set globally via terminal.backend in config.yaml (mapped to TERMINALENV). You pick local, ssh, docker, modal, daytona, or singularity, and every terminal() call routes…","versions":[],"error_snippets":["timeout: 180","\"description\": \"Execution target: 'local' (default) or a named backend (e.g. 'macbook'). Omit for default.\"","doesn't persist. Environment variables don't survive. Shell state is gone between calls. The agent ends up prepending","terminal_tool.py:463","terminal_tool.py:539"],"workaround":null,"fixed_in":null,"score":119.1,"first_seen":"2026-10-18T04:29:32Z"},{"id":"openclaw/openclaw#38327","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":38327,"title":"[Bug] \"Cannot convert undefined or null to object\" in 2026.3.2 with google-vertex/gemini-3.1-pro-preview","url":"https://github.com/openclaw/openclaw/issues/38327","state":"open","state_reason":null,"labels":["bug","regression","P1","clawsweeper:no-new-fix-pr","clawsweeper:fix-shape-clear","clawsweeper:needs-maintainer-review","clawsweeper:source-repro","impact:auth-provider","issue-rating: 🦞 diamond lobster"],"created":"2026-03-06T20:30:59Z","updated":"2026-10-22T14:32:43Z","closed":null,"reactions":4,"thumbs_up":3,"comments":18,"severity":"high","severity_reasons":["label:p1"],"topic":"install/update","classes":["install/update","auth/session","network"],"cosmetic_ui":false,"summary":"Regression (worked before, now fails) After updating to 2026.3.2, any message causes embedded agent to fail with \"Cannot convert undefined or null to object\". Update to 2026.3.2 (npm install -g openclaw@latest) Set model to…","versions":["2026.3.2","2026.2.26"],"error_snippets":["Mar 07 01:06:49 ... [agent/embedded] embedded run agent end: runId=... isError=true error=Cannot convert undefined or null to object","Mar 07 01:23:08 ... [agent/embedded] embedded run agent end: runId=... isError=true error=Cannot convert undefined or null to object","After updating to 2026.3.2, any message causes embedded agent to fail with \"Cannot convert undefined or null to object\".","embedded run agent end: ... error=Cannot convert undefined or null to object","Also seeing occasional 429 rate limit on embeddings"],"workaround":null,"fixed_in":null,"score":119.0,"first_seen":"2026-10-18T04:30:03Z"},{"id":"NousResearch/hermes-agent#142461","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":142461,"title":"Automation wakes into auto-created Discord threads are refused admission_conflict forever (parents never resume after children finish)","url":"https://github.com/NousResearch/hermes-agent/issues/142461","state":"open","state_reason":null,"labels":["type/bug","comp/gateway","tool/delegate","platform/discord","P3","area/sessions"],"created":"2026-10-21T08:54:10Z","updated":"2026-10-25T13:36:04Z","closed":null,"reactions":6,"thumbs_up":6,"comments":8,"severity":"medium","severity_reasons":["class:always-on/daemon"],"topic":"always-on/daemon","classes":["always-on/daemon","install/update"],"cosmetic_ui":false,"summary":"With discord.autothread: true, every automation wake that targets a session whose thread was auto-created is refused at admission: The drain requeues the event and retries forever, so parent agents never resume after their children finish…","versions":["0.22.2-era"],"error_snippets":["restored.source (latest in-thread envelope): {platform, chat_id, chat_type, thread_id, user_id, ...}","event.source (origin-rebuilt): {... same 12 keys ..., auto_thread_created: True, auto_thread_initial_name: '...'}","With `discord.auto_thread: true`, every automation wake that targets a session whose thread was **auto-created** is refused at admission:","The drain requeues the event and retries forever, so **parent agents never resume after their children finish**: `delegate_task` children complete (results verifiable in their transcripts), background-process watches fir","Reproduced deterministically on a live gateway (v0.22.2-era checkout, single profile, single Discord adapter): refusals from the first attempt, across gateway restarts and code updates, on every auto-created-thread sessi"],"workaround":null,"fixed_in":null,"score":118.4,"first_seen":"2026-10-23T09:23:24Z"},{"id":"omacom/omarchy#6223","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":6223,"title":"Crash and restart after omarchy-system-lock","url":"https://github.com/omacom/omarchy/issues/6223","state":"open","state_reason":null,"labels":["bug","compatibility"],"created":"2026-07-14T20:38:58Z","updated":"2026-10-08T03:22:08Z","closed":null,"reactions":5,"thumbs_up":5,"comments":27,"severity":"high","severity_reasons":["title:breakage"],"topic":"crash/hang","classes":["crash/hang","install/update","network"],"cosmetic_ui":false,"summary":"Omarchy 3.8.3 PC: Framework Desktop (AMD Ryzen AI Max 300 Series) CPU/APU: AMD Ryzen AI Max 385 w/ Radeon 8050S (Strix Halo, gfx11) GPU: amdgpu 0000:c2:00.0, VRAM 16384M (UMA), ATOM BIOS 113-STRXLGEN-001 SMU firmware: 0x0a640600 (100.6.0)…","versions":["3.8.3","00.0","100.6.0","7.1.3-arch1"],"error_snippets":["--- journal ends, next boot starts ~1 min later, no panic/reset/coredump anywhere ---","There is no clean shutdown in the journal (no unmounting sequence, no kernel panic, no amdgpu reset/timeout messages) — the journal just stops, consistent with a hard hang forcing an external/hardware reset.","Journal simply ends mid-stream, no error, no shutdown sequence:"],"workaround":{"text":"I changed only `blankProcess` in `Service.qml` from:","source":"comment by @aledeniz (1 reactions)","url":"https://github.com/omacom/omarchy/issues/6223#issuecomment-5362858718"},"fixed_in":null,"score":115.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"anthropics/claude-code#98256","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":98256,"title":"[BUG] MCP elicitation: capability is declared but requests are auto-declined (\"print mode\") in interactive VSCode sessions","url":"https://github.com/anthropics/claude-code/issues/98256","state":"open","state_reason":null,"labels":["bug","has repro","api:bedrock","platform:macos","area:mcp","platform:vscode"],"created":"2026-09-29T21:22:42Z","updated":"2026-10-16T03:15:29Z","closed":null,"reactions":16,"thumbs_up":16,"comments":7,"severity":"medium","severity_reasons":["engagement 23","class:mcp/tools"],"topic":"mcp/tools","classes":["mcp/tools","self-host/headless"],"cosmetic_ui":false,"summary":"In an interactive VSCode-extension session, Claude Code declares the MCP elicitation capability during initialize, but auto-declines every elicitation/create request without showing any UI. The same session can render permission prompts…","versions":["2.1.284"],"error_snippets":["In an interactive VSCode-extension session, Claude Code declares the MCP elicitation capability during initialize, but auto-declines every elicitation/create request without showing any UI. The same session can render pe","Because the capability is declared, servers cannot detect the situation:","1. Minimal Python MCP stdio server (FastMCP / mcp SDK) with a tool that calls ctx.elicit(message=..., schema=...).","2. Register it in .mcp.json, run a Claude Code session from the VSCode extension.","4. The tool returns immediately with a non-accept elicitation result; no dialog is shown."],"workaround":null,"fixed_in":null,"score":114.9,"first_seen":"2026-10-19T18:38:33Z"},{"id":"anthropics/claude-code#102412","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":102412,"title":"[BUG] Unable to switch off automatic worktree creation","url":"https://github.com/anthropics/claude-code/issues/102412","state":"open","state_reason":null,"labels":["bug","platform:windows","regression","area:desktop"],"created":"2026-10-15T13:24:23Z","updated":"2026-10-22T15:50:21Z","closed":null,"reactions":9,"thumbs_up":9,"comments":9,"severity":"high","severity_reasons":["label:regression"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"Am I the only one no longer able to figure out how to switch off automatic work tree creation in the newest version. Updated to the latest claude code to do development on the desktop app and now it seems I'm forced into using worktrees?…","versions":["2.1.206"],"error_snippets":["Let me easily pick if I want to use worktrees","2. Start up a new session - unable to figure out how to use a worktree or not"],"workaround":null,"fixed_in":null,"score":113.9,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openclaw/openclaw#158390","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":158390,"title":"plugin-captures tmp dirs not GC'd after build/catalog operations — disk fills indefinitely","url":"https://github.com/openclaw/openclaw/issues/158390","state":"open","state_reason":null,"labels":["no-stale","P1","clawsweeper:fix-shape-clear","clawsweeper:queueable-fix","clawsweeper:source-repro","impact:crash-loop","issue-rating: 🦞 diamond lobster"],"created":"2026-09-25T22:18:19Z","updated":"2026-10-24T09:38:56Z","closed":null,"reactions":2,"thumbs_up":1,"comments":18,"severity":"critical","severity_reasons":["label:crash-loop"],"topic":"crash/hang","classes":["crash/hang","mcp/tools","always-on/daemon"],"cosmetic_ui":false,"summary":"/home/ /.openclaw/tmp/plugin-captures/ accumulates multi-GB session directories from plugin-build and model-catalog operations. These directories are never cleaned up after the operation completes, so disk usage grows permanently with each…","versions":[],"error_snippets":["├── 9611f0b0-... 2.2G ← orphaned (empty owner.sqlite, no process handles)","lsof +D ~/.openclaw/tmp/plugin-captures/<dir> 2>/dev/null | wc -l","`/home/ /.openclaw/tmp/plugin-captures/` accumulates multi-GB session directories from plugin-build and model-catalog operations. These directories are never cleaned up after the operation completes, so disk usage grows ","Each directory contains:","Completed build/catalog session dirs should be deleted (or at minimum their `node_modules` pruned) once the operation finishes and the owning gateway process exits."],"workaround":{"text":"Manually delete orphaned dirs where `owner.sqlite` is empty and no process holds a file handle:","source":"issue body","url":"https://github.com/openclaw/openclaw/issues/158390"},"fixed_in":null,"score":111.8,"first_seen":"2026-10-18T04:33:05Z"},{"id":"openai/codex#47345","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":47345,"title":"Bug: Linux sandbox fails with mountinfo path is not absolute after ChatGPT update","url":"https://github.com/openai/codex/issues/47345","state":"closed","state_reason":"completed","labels":["bug","sandbox","CLI","app","Linux"],"created":"2026-09-22T19:13:30Z","updated":"2026-10-15T18:50:08Z","closed":"2026-09-24T16:27:49Z","reactions":37,"thumbs_up":37,"comments":52,"severity":"medium","severity_reasons":["engagement 89","class:install/update"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"26.917.51856 Ubuntu 24.04.5 LTS (Noble) After the ChatGPT Linux application updated from 26.915.31029 to 26.917.51856, Codex can no longer execute commands using its normal Linux sandbox. The failure is reproducible directly using the…","versions":["26.917.51856","26.915.31029","0.9.0","0.155.0-alpha.16"],"error_snippets":["error building bubblewrap command: mountinfo path is not absolute","After the ChatGPT Linux application updated from 26.915.31029 to 26.917.51856, Codex can no longer execute commands using its normal Linux sandbox.","The failure is reproducible directly using the Codex CLI bundled with the ChatGPT application, independently of the GUI or any particular repository.","Exit status is 1.","Within a normal Codex agent session, sandboxed shell commands and apply_patch fail with the same underlying error. For example, the agent reports:"],"workaround":null,"fixed_in":null,"score":111.4,"first_seen":"2026-10-18T04:31:06Z"},{"id":"openai/codex#49365","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":49365,"title":"Codex Desktop: exec_command fails in WSL-style session with os error 2, including pwd in /tmp, after restart","url":"https://github.com/openai/codex/issues/49365","state":"open","state_reason":null,"labels":["bug","windows-os","tool-calls","app"],"created":"2026-09-29T21:01:10Z","updated":"2026-10-16T13:52:49Z","closed":null,"reactions":13,"thumbs_up":13,"comments":10,"severity":"medium","severity_reasons":["engagement 23","class:self-host/headless"],"topic":"self-host/headless","classes":["self-host/headless","auth/session"],"cosmetic_ui":false,"summary":"Windows package version observed in app logs: 26.928.1915.0. Desktop client version reported in runtime logs: 26.928.20755. Bundled WSL CLI verified with --version: codex-cli 0.159.0. Not provided. Codex Desktop on Windows, with the agent…","versions":["26.928.20755","0.159.0"],"error_snippets":["exec_command failed: CreateProcess { message: \"Rejected(\\\"Failed to create unified exec process: No such file or directory (os error 2)\\\")\" }","bundled_executable_relocation_failed","errorCode=ENOENT","errorMessage=\"Bundled executable relocation failed during stat_source\"","plugin_marketplace_add_failed"],"workaround":null,"fixed_in":null,"score":110.6,"first_seen":"2026-10-18T04:34:23Z"},{"id":"anthropics/claude-code#102677","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":102677,"title":"Desktop (Code tab): scheduled task runs no longer visible in the sidebar","url":"https://github.com/anthropics/claude-code/issues/102677","state":"open","state_reason":null,"labels":["bug","platform:macos","regression","area:desktop","area:routines"],"created":"2026-10-16T08:13:08Z","updated":"2026-10-25T12:31:21Z","closed":null,"reactions":4,"thumbs_up":4,"comments":4,"severity":"high","severity_reasons":["label:regression"],"topic":"always-on/daemon","classes":["always-on/daemon","install/update","crash/hang"],"cosmetic_ui":false,"summary":"After updating Claude Desktop, the Routines section disappeared from the Code tab sidebar. Previously every scheduled task run appeared there, so I could see at a glance which tasks ran, when, and whether they succeeded. Now there is no…","versions":[],"error_snippets":["Previously every scheduled task run appeared there, so I could see at a glance which","tasks ran, when, and whether they succeeded. Now there is no Routines entry in the sidebar,","infra and dev digests, weekly reports). The only way to check them now is to open each","\"did everything run this morning?\"."],"workaround":null,"fixed_in":null,"score":110.4,"first_seen":"2026-10-21T09:24:21Z"},{"id":"github/copilot-cli#4851","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":4851,"title":"Azure MCP server fails sending HTTP request","url":"https://github.com/github/copilot-cli/issues/4851","state":"open","state_reason":null,"labels":["triage"],"created":"2026-09-15T09:40:38Z","updated":"2026-10-01T21:31:12Z","closed":null,"reactions":10,"thumbs_up":10,"comments":5,"severity":"medium","severity_reasons":["engagement 15","class:self-host/headless","class:mcp/tools"],"topic":"self-host/headless","classes":["self-host/headless","mcp/tools","network"],"cosmetic_ui":false,"summary":"This broke overnight. We've been using the same Azure MCP registry for months fine. Copilot CLI Rust runtime fails with BrokenPipe when validating Azure API Center MCP registry. GitHub Copilot CLI 1.0.83 cannot validate MCP servers against…","versions":["1.0.83","1.2","1.1","0.13.4"],"error_snippets":["[rust:hyper_util::client::legacy::connect::http] connected to ...:443","[rust:hyper_util::client::legacy::client] client connection error:","hyper::Error(Shutdown, Os {","Failed to load registry: Error: HTTP request failed: error sending request for url ( <redacted> )","GitHub Copilot CLI 1.0.83 cannot validate MCP servers against an Azure API Center registry when enterprise MCP access is configured as Registry only."],"workaround":null,"fixed_in":null,"score":110.2,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openai/codex#48422","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":48422,"title":"Windows: visible console windows flash for shell process children on every session/turn","url":"https://github.com/openai/codex/issues/48422","state":"closed","state_reason":"duplicate","labels":["bug","windows-os","CLI","app-server"],"created":"2026-09-26T10:30:37Z","updated":"2026-09-29T06:55:59Z","closed":"2026-09-29T06:55:59Z","reactions":31,"thumbs_up":31,"comments":30,"severity":"medium","severity_reasons":["engagement 61","class:self-host/headless"],"topic":"self-host/headless","classes":["self-host/headless","always-on/daemon","network"],"cosmetic_ui":false,"summary":"codex-cli 0.157.1 Microsoft Windows NT 10.0.26200.0 x64 Windows Terminal + Powershell On Windows, a batch of visible console windows flashes for git.exe / pwsh.exe (and other helper) child processes: several git.exe windows on every new…","versions":["0.157.1"],"error_snippets":["(session start + every turn) and","codex-rs/tui/src/startup_orchestration.rs:494-544","codex-rs/core/src/turn_metadata.rs:547-551","codex-rs/exec-server/src/shell_snapshot.rs:287","codex-rs/hooks/src/engine/command_runner.rs:233-245"],"workaround":null,"fixed_in":null,"score":109.9,"first_seen":"2026-10-19T18:38:33Z"},{"id":"github/copilot-cli#5197","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":5197,"title":"Voice server segfaults (signal 11) on WSL2 after updating to 1.0.100; /voice stuck on 'Fetching voice models'","url":"https://github.com/github/copilot-cli/issues/5197","state":"open","state_reason":null,"labels":["area:input-keyboard"],"created":"2026-10-19T22:28:49Z","updated":"2026-10-22T21:59:08Z","closed":null,"reactions":2,"thumbs_up":2,"comments":4,"severity":"medium","severity_reasons":["class:self-host/headless","class:install/update"],"topic":"self-host/headless","classes":["self-host/headless","install/update","crash/hang","network"],"cosmetic_ui":false,"summary":"After updating to Copilot CLI 1.0.100 (which also pulled a new voice runtime), voice input no longer works on WSL2. The voice server process segfaults (signal 11) a few seconds after starting, so /voice stays on \"Fetching voice models\" and…","versions":["1.0.100","1.0.98"],"error_snippets":["[INFO] voice server: boot pid=1708 endpoint=/run/user/1000/copilot-voice/<id>.sock","[WARNING] [VoiceEngine] connection error: read ECONNRESET","[ERROR] [VoiceEngine] fatal: Engine connection closed.","[WARNING] [VoiceEngine] start failed: Engine connection closed.","copilot: MainThread: potentially unexpected fatal signal 11."],"workaround":null,"fixed_in":null,"score":106.5,"first_seen":"2026-10-20T09:28:14Z"},{"id":"openai/codex#53891","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":53891,"title":"[macOS] Update checks fail because appcast endpoint returns HTTP 520","url":"https://github.com/openai/codex/issues/53891","state":"closed","state_reason":"completed","labels":["bug","app","connectivity"],"created":"2026-10-14T06:42:32Z","updated":"2026-10-21T17:29:43Z","closed":"2026-10-21T17:29:43Z","reactions":17,"thumbs_up":17,"comments":40,"severity":"medium","severity_reasons":["engagement 57","class:install/update"],"topic":"install/update","classes":["install/update","self-host/headless"],"cosmetic_ui":false,"summary":"26.1007.71326 (build 20398), production channel Updater reports plantype=promax Darwin 27.2.0 arm64 arm — macOS 27.2 (26B5112a), Apple Silicon Selecting Check for Updates shows: The failure happens while fetching the update feed, before an…","versions":["26.1007.71326","27.2.0"],"error_snippets":["A network error occurred while downloading https://updates.oaistatic.com/codex/app/appcast?installation_id=<redacted>&arch=arm64&app_version=26.1007.71326&beta=false&os-version=27.2.0&plan_type=promax&codex_cache_bust=4.","SUSparkleErrorDomain Code=2001","status=error","error code: 520","enableSparkle=true"],"workaround":null,"fixed_in":null,"score":106.1,"first_seen":"2026-10-22T09:27:06Z"},{"id":"openclaw/openclaw#84110","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":84110,"title":"[Bug]: Codex app-server rewrites prompt on tool-call continuation turns, busting OpenAI prompt cache mid-turn (cache ratio 93% → 47%)","url":"https://github.com/openclaw/openclaw/issues/84110","state":"open","state_reason":null,"labels":["bug","stale","P2","clawsweeper:no-new-fix-pr","clawsweeper:needs-maintainer-review","clawsweeper:needs-info","impact:session-state","impact:auth-provider","issue-rating: 🦪 silver shellfish"],"created":"2026-05-19T11:36:59Z","updated":"2026-10-23T15:55:28Z","closed":null,"reactions":5,"thumbs_up":2,"comments":9,"severity":"medium","severity_reasons":["class:self-host/headless","class:auth/session"],"topic":"self-host/headless","classes":["self-host/headless","auth/session","install/update","mcp/tools"],"cosmetic_ui":false,"summary":"Regression (worked before, now fails) Since OC v2026.5.12 + codex-cli 0.130.0 (first observed 2026-05-15), the Codex app-server rewrites the full prompt between a user turn and its tool-call continuation turn instead of appending only the…","versions":["2026.5.12","0.130.0","3.5"],"error_snippets":["Turn 2 (exec tool call): cached=1","Turn 5 (exec tool call): cached=0","Mid-turn bust rate post-update: 100% of turns following custom_tool_call name=exec","and its tool-call continuation turn instead of appending only the tool_call_output, invalidating the OpenAI prompt cache on every exec","tool call and raising effective per-token costs by ~3.5×."],"workaround":null,"fixed_in":null,"score":105.9,"first_seen":"2026-10-19T18:38:33Z"},{"id":"anthropics/claude-code#93537","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":93537,"title":"macOS: concurrent claude processes zero MCP OAuth entries in shared Keychain blob despite valid refresh tokens","url":"https://github.com/anthropics/claude-code/issues/93537","state":"open","state_reason":null,"labels":["bug","has repro","platform:macos","area:auth","area:mcp","area:agent-view"],"created":"2026-09-11T04:52:54Z","updated":"2026-10-22T00:24:52Z","closed":null,"reactions":4,"thumbs_up":4,"comments":5,"severity":"medium","severity_reasons":["class:auth/session","class:mcp/tools"],"topic":"auth/session","classes":["auth/session","mcp/tools","network","self-host/headless","install/update"],"cosmetic_ui":false,"summary":"On macOS, MCP OAuth entries in the shared Keychain item Claude Code-credentials get rewritten with empty accessToken/refreshToken (and no clientId) while the refresh tokens are still valid, when several claude processes touch the store…","versions":["2.1.268","2.1.206","2.1.263","2.1.267"],"error_snippets":["and re-authenticate every server on every new session, even seconds after authenticating in another terminal. Related but distinct from #91009 (Windows, file store,","stub blob (same keys, empty tokens) written to the fallback","entries now have empty","unset everywhere, login keychain unlocked with no timeout. User history shows 33","No `/logout`, no \"Clear authentication\", `CLAUDE_CONFIG_DIR` unset everywhere, login keychain unlocked with no timeout. User history shows 33 `/mcp` re-auths over two months, clustered at session starts."],"workaround":null,"fixed_in":null,"score":105.4,"first_seen":"2026-10-22T09:27:06Z"},{"id":"openclaw/openclaw#159662","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":159662,"title":"prepared-model-catalog.worker.js: unbounded memory leak, ~4-5 GB/h, provider-agnostic (reproduced on cold reboot + provider bisect)","url":"https://github.com/openclaw/openclaw/issues/159662","state":"open","state_reason":null,"labels":["impact:crash-loop","P0","issue-rating: 🦪 silver shellfish"],"created":"2026-09-27T13:12:30Z","updated":"2026-10-21T09:36:35Z","closed":null,"reactions":2,"thumbs_up":1,"comments":24,"severity":"critical","severity_reasons":["label:p0"],"topic":"crash/hang","classes":["crash/hang","always-on/daemon","install/update","mcp/tools"],"cosmetic_ui":false,"summary":"The gateway's prepared-model-catalog.worker.js worker thread leaks memory continuously and monotonically, independent of workload volume, model provider, or plugin configuration. On an idle-to-light-traffic single-user install, RSS grows…","versions":["2.5","2026.9.6","1.72","1.42"],"error_snippets":["openclaw gateway diagnostics export","openclaw gateway call diagnostics.heapProfile --timeout 30000","agents.entries.main.model","node_modules/@openclaw/ai/dist/event-stream-*.mjs:38","node_modules/undici/lib/web/fetch/response.js:548"],"workaround":null,"fixed_in":null,"score":105.2,"first_seen":"2026-10-18T04:33:05Z"},{"id":"omacom/omarchy#12436","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":12436,"title":"Deletion of master branch breaks updates + manual install","url":"https://github.com/omacom/omarchy/issues/12436","state":"open","state_reason":null,"labels":["bug"],"created":"2026-09-18T17:17:23Z","updated":"2026-10-13T16:41:59Z","closed":null,"reactions":15,"thumbs_up":15,"comments":10,"severity":"medium","severity_reasons":["engagement 25","class:install/update"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"The master branch in this repo was deleted apparently around September 14th. This breaks the omarchy-update-git step for any system where the local git copy is set up to track the master branch. Since the documented flow for upgrading to…","versions":[],"error_snippets":[],"workaround":null,"fixed_in":null,"score":104.6,"first_seen":"2026-10-19T18:38:33Z"},{"id":"anthropics/claude-code#93562","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":93562,"title":"Egress allowlist ignores \"All domains\": every host outside a built-in list gets 403 on CONNECT (regression 2026-09-10 -> 2026-09-11)","url":"https://github.com/anthropics/claude-code/issues/93562","state":"open","state_reason":null,"labels":["bug","area:cowork","regression","area:networking","area:routines"],"created":"2026-09-11T07:39:06Z","updated":"2026-09-14T13:20:59Z","closed":null,"reactions":6,"thumbs_up":6,"comments":8,"severity":"high","severity_reasons":["label:regression","title:breakage"],"topic":"install/update","classes":["install/update","auth/session","always-on/daemon","self-host/headless","network"],"cosmetic_ui":false,"summary":"With Allow network egress = ON and Domain allowlist = \"All domains\" (Settings -> Capabilities -> Code execution), the session egress proxy still rejects every host outside a small built-in allowlist with 403 on CONNECT. This is a…","versions":[],"error_snippets":["$ curl -sS -o /dev/null -w '%{http_code}\\n' https://json2news.radiobutler.com/","curl: (56) CONNECT tunnel failed, response 403","$ curl -sS -o /dev/null -w '%{http_code}\\n' https://news.google.com/","$ curl -sS -o /dev/null -w '%{http_code}\\n' https://example.com/","$ curl -sS -o /dev/null -w '%{http_code}\\n' https://api.github.com/"],"workaround":null,"fixed_in":null,"score":104.1,"first_seen":"2026-10-18T04:33:44Z"},{"id":"github/copilot-cli#1803","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":1803,"title":"Support MCP resources/read primitive","url":"https://github.com/github/copilot-cli/issues/1803","state":"open","state_reason":null,"labels":["area:mcp"],"created":"2026-03-03T13:57:56Z","updated":"2026-10-05T16:17:01Z","closed":null,"reactions":14,"thumbs_up":14,"comments":2,"severity":"medium","severity_reasons":["engagement 16","class:mcp/tools"],"topic":"mcp/tools","classes":["mcp/tools","self-host/headless","auth/session"],"cosmetic_ui":false,"summary":"MCP servers can expose data through the resources primitive (resources/list, resources/read). This is one of the three core MCP primitives alongside tools and prompts. Currently, Copilot CLI only supports MCP tools. When an MCP tool…","versions":["2.2.2"],"error_snippets":["✓ search_topics: pekko kafka consumer actor ordering sharding entity","└ Found 5 results for \"pekko kafka consumer actor ordering sharding entity\":","TypeError: fetch failed","MCP servers can expose data through the `resources` primitive (`resources/list`, `resources/read`). This is one of the three core MCP primitives alongside tools and prompts."],"workaround":null,"fixed_in":null,"score":104.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"anthropics/claude-code#96718","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":96718,"title":"[BUG] Artifact \"Version history\" removed from viewer menu across Claude Code, Cowork and claude.ai – saved versions unreachable (related to #95442)","url":"https://github.com/anthropics/claude-code/issues/96718","state":"open","state_reason":null,"labels":["bug","platform:windows","regression"],"created":"2026-09-24T12:04:20Z","updated":"2026-10-22T07:56:15Z","closed":null,"reactions":13,"thumbs_up":13,"comments":9,"severity":"high","severity_reasons":["label:regression"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"Published artifacts no longer show a \"Version history\" option anywhere: not in the claude.ai web viewer's title menu, and not in the Claude desktop app's artifact side panel. Earlier versions are still being saved (each publish returns a…","versions":[],"error_snippets":["No error message. The menu option is simply missing.","Published artifacts no longer show a \"Version history\" option anywhere: not in the claude.ai web viewer's title menu, and not in the Claude desktop app's artifact side panel. Earlier versions are still being saved (each ","Version history is the only way for an owner to see and open earlier versions. Without it, saved versions are unreachable: you can't even see which versions exist to ask for one. Asking Claude to republish an older versi","A \"Version history\" entry (previously shown as \"Version history (count)\" between Share and Refresh) that lists every published version and lets the owner open any of them. The docs still describe this: \"Each publish beco","Please restore Version history in the web viewer and the desktop app, for Claude Code, Cowork and claude.ai artifacts, and confirm that earlier versions of existing artifacts are retained. If this was changed on purpose,"],"workaround":null,"fixed_in":null,"score":103.8,"first_seen":"2026-10-18T04:33:44Z"},{"id":"NousResearch/hermes-agent#141862","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":141862,"title":"Plugin tools can't declare execution_timeout — register_tool() drops it, so every plugin tool is capped at the 420s deadline","url":"https://github.com/NousResearch/hermes-agent/issues/141862","state":"open","state_reason":null,"labels":["type/bug","comp/cli","comp/plugins","P2"],"created":"2026-10-20T04:04:01Z","updated":"2026-10-20T05:42:41Z","closed":null,"reactions":4,"thumbs_up":2,"comments":7,"severity":"medium","severity_reasons":["class:mcp/tools","class:network"],"topic":"mcp/tools","classes":["mcp/tools","network"],"cosmetic_ui":false,"summary":"Plugin tools can't raise their own execution deadline. PluginContext.registertool() (hermescli/plugins.py) does not accept or forward executiontimeout, but tools.registry.register() supports it and…","versions":[],"error_snippets":["execution_timeout","agent.tool_dispatch_helpers._extend_tool_timeout()","_DEFAULT_CONCURRENT_TOOL_TIMEOUT_S","timeouts.tools.sequential_call","subprocess.run(..., timeout=900)"],"workaround":{"text":"`timeouts.tools.sequential_call: 960` in `config.yaml`. This raises the ceiling for every tool, so a wedged tool now blocks for up to 16 minutes instead of 7. It also leaves the concurrent-batch path at 420 s.","source":"issue body","url":"https://github.com/NousResearch/hermes-agent/issues/141862"},"fixed_in":null,"score":103.0,"first_seen":"2026-10-20T09:28:14Z"},{"id":"openai/codex#49731","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":49731,"title":"Windows app with \"Run agent in WSL\": every command fails with \"Failed to create unified exec process: No such file or directory\" (arg0 helper dir deleted by Windows exec-server)","url":"https://github.com/openai/codex/issues/49731","state":"closed","state_reason":"completed","labels":["bug","windows-os","tool-calls","app"],"created":"2026-09-30T20:14:25Z","updated":"2026-10-16T06:54:29Z","closed":"2026-10-12T15:37:22Z","reactions":27,"thumbs_up":27,"comments":47,"severity":"medium","severity_reasons":["engagement 74","class:self-host/headless"],"topic":"self-host/headless","classes":["self-host/headless"],"cosmetic_ui":false,"summary":"Powered by Codex & OWL Version 26.928.21956 Microsoft Windows NT 10.0.26200.0 x64 no-active-thread-01a0f3f0-94ff-7451-83bf-3238585cc8fe Subject: Codex desktop on Windows with \"Run agent in WSL\" enabled: every command fails with \"Failed to…","versions":["26.928.21956","10.0.26200"],"error_snippets":["Subject: Codex desktop on Windows with \"Run agent in WSL\" enabled: every command fails with \"Failed to create unified exec process: No such file or directory\"","Every command the agent tries to run (git, pwd, with any cwd or shell) fails before execution:","exec_command failed: CreateProcess { message: \"Rejected(\\\"Failed to create unified exec process: No such file or directory (os error 2)\\\")\" }","What I found (strace -f -p , filtered on ENOENT)","execve(\".../tmp/arg0/codex-arg0XXXXXX/codex-linux-sandbox\", [... \"--command-cwd\", \"/tmp\", ... \"--\", \"/bin/bash\", \"-c\", \"pwd\"]) = -1 ENOENT"],"workaround":null,"fixed_in":null,"score":102.7,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openai/codex#48125","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":48125,"title":"I CANT FUCKING COPY TEXT, IT WAS WORKING JUST FINE MANI CANT FUCKING COPY TEXT, IT WAS WORKING JUST FINE MAN","url":"https://github.com/openai/codex/issues/48125","state":"closed","state_reason":"not_planned","labels":["bug","TUI","CLI","remote"],"created":"2026-09-25T15:57:48Z","updated":"2026-10-15T09:03:40Z","closed":"2026-09-29T03:19:54Z","reactions":45,"thumbs_up":31,"comments":24,"severity":"medium","severity_reasons":["engagement 69","class:self-host/headless"],"topic":"self-host/headless","classes":["self-host/headless"],"cosmetic_ui":false,"summary":"ubuntu 24.04 using ssh to it ubuntu default terminal i cant fucking copy text, the reason codex-cli is the best was because i could fucking copy text normally from my cli , and it was feeling like a cli not like a fucking GUI application…","versions":["1.157.0"],"error_snippets":[],"workaround":null,"fixed_in":null,"score":101.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"google-gemini/gemini-cli#21409","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":21409,"title":"Generalist agent hangs","url":"https://github.com/google-gemini/gemini-cli/issues/21409","state":"open","state_reason":null,"labels":["priority/p1","area/agent","🔒 maintainer only","workstream-rollup","status/need-retesting","status/bot-triaged","kind/bug"],"created":"2026-03-06T15:40:32Z","updated":"2026-10-26T09:26:04Z","closed":null,"reactions":8,"thumbs_up":8,"comments":8,"severity":"high","severity_reasons":["label:p1","title:breakage"],"topic":"crash/hang","classes":["crash/hang","auth/session"],"cosmetic_ui":false,"summary":"Hey! Whenever gemini-cli defers to the generalist agent it hangs forever. Simple changes like folder creation hang. I've let it wait for up to an hour before cancelling it. Instructing the model not to defer to sub agents resolves this…","versions":["0.32.1"],"error_snippets":["to enter the interactive CLI, then run the","Run `gemini` to enter the interactive CLI, then run the `/about` command."],"workaround":null,"fixed_in":null,"score":101.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openai/codex#49618","harness":"codex","harness_name":"Codex CLI","repo":"openai/codex","number":49618,"title":"Codex Remote pairing loop between Windows and Android — “Approve this phone” repeats","url":"https://github.com/openai/codex/issues/49618","state":"closed","state_reason":"completed","labels":["bug","windows-os","auth","app","remote"],"created":"2026-09-30T13:04:37Z","updated":"2026-10-07T08:53:19Z","closed":"2026-10-06T03:33:14Z","reactions":16,"thumbs_up":16,"comments":29,"severity":"medium","severity_reasons":["engagement 45","class:self-host/headless","class:auth/session"],"topic":"self-host/headless","classes":["self-host/headless","auth/session"],"cosmetic_ui":false,"summary":"I’m experiencing a Windows ↔ Android Remote pairing failure. Windows ChatGPT desktop: 26.904.11930 Android ChatGPT: 1.2026.265 ChatGPT Plus / Personal workspace Same account and workspace on both devices Windows generates the QR normally…","versions":["26.904.11930"],"error_snippets":["I’m experiencing a Windows ↔ Android Remote pairing failure.","This appears to be a controller-claim/device-binding failure because the Windows host QR never refreshes and the allowed-device list remains empty after mobile approval."],"workaround":null,"fixed_in":null,"score":100.6,"first_seen":"2026-10-19T18:38:33Z"},{"id":"anthropics/claude-code#97319","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":97319,"title":"[BUG] MCP client rejects valid tools/list response due to strict validation of ttlMs/cacheScope fields (Roblox Studio MCP server)","url":"https://github.com/anthropics/claude-code/issues/97319","state":"open","state_reason":null,"labels":["bug","has repro","platform:windows","area:mcp"],"created":"2026-09-26T01:07:59Z","updated":"2026-09-27T13:37:10Z","closed":null,"reactions":6,"thumbs_up":6,"comments":11,"severity":"medium","severity_reasons":["engagement 17","class:self-host/headless","class:mcp/tools"],"topic":"self-host/headless","classes":["self-host/headless","mcp/tools"],"cosmetic_ui":false,"summary":"A user-configured stdio MCP server connects successfully (status: connected) but always reports toolcount: 0, across brand-new sessions, app restarts, and even a full OS reboot. The problem persists indefinitely — no retry ever recovers…","versions":[],"error_snippets":["Connection established with capabilities: {\"hasTools\":true, ... \"negotiatedProtocolVersion\":\"2026-07-28\"}","tools/list failed (Invalid result for tools/list: [","Failed to fetch tools: Invalid result for tools/list: [...]","The server's tools/list response apparently includes extra per-tool/result metadata fields (ttlMs, cacheScope) tied to a newer MCP protocol revision. The Claude Code MCP client's schema validator rejects the whole respon","Expected behavior"],"workaround":null,"fixed_in":null,"score":99.9,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openclaw/openclaw#156571","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":156571,"title":"[Bug]: 2026.9.5 model-catalog worker leaks openclaw-plugin-build-* source captures in tmp (1-3 GB/min, fills the disk) — disk side of #155753","url":"https://github.com/openclaw/openclaw/issues/156571","state":"open","state_reason":null,"labels":["clawsweeper:needs-info","impact:crash-loop","P0","issue-rating: 🦐 gold shrimp","impact:other","clawsweeper:manual-only"],"created":"2026-09-23T14:30:34Z","updated":"2026-10-09T06:19:09Z","closed":null,"reactions":2,"thumbs_up":1,"comments":15,"severity":"critical","severity_reasons":["label:p0"],"topic":"crash/hang","classes":["crash/hang","mcp/tools","always-on/daemon","install/update"],"cosmetic_ui":false,"summary":"Regression (worked before, now fails) On 2026.9.5 the prepared model-catalog worker keeps creating plugin source captures under $OPENCLAWHOME/tmp/openclaw-model-catalog-/openclaw-plugin-build- and never deletes them. On our gateway this…","versions":["2026.9.5","2.8","58.8","78.7"],"error_snippets":["provider catalog refresh failed: Cannot find module '.../tmp/openclaw-model-catalog-XXXX/openclaw-plugin-build-GrYv8x/package-0/node_modules/llama-cpp/dist/provider-policy-api.js'","provider catalog refresh failed: Cannot find module .../openclaw-plugin-build-XXXX/package-0/node_modules/llama-cpp/dist/provider-policy-api.js","plugins.entries.codex.config.sessionCatalog.enabled=false","This is the disk-side symptom of the same worker loop reported for CPU in #155753 (see also #154276, #153422): the gateway also sits at 100-260% CPU. We are filing it separately because the disk leak is what actually tak","2. Start the gateway and leave it idle (no client connected on the gateway port, `tasks` empty)."],"workaround":{"text":"a cron every minute that, inside each `openclaw-model-catalog-` folder, keeps the 10 oldest captures (the ones created at worker start) and deletes the rest older than 1 minute, by creation time (mtime keeps changing).","source":"issue body","url":"https://github.com/openclaw/openclaw/issues/156571"},"fixed_in":null,"score":99.6,"first_seen":"2026-10-18T04:33:05Z"},{"id":"github/copilot-cli#4905","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":4905,"title":"Desktop app: sessions die minutes after spawn — \"GitHub credential registration is no longer available for this session\" makes the github-mcp-server catalog stale and fatal","url":"https://github.com/github/copilot-cli/issues/4905","state":"open","state_reason":null,"labels":["triage"],"created":"2026-09-18T20:03:30Z","updated":"2026-09-27T17:44:04Z","closed":null,"reactions":4,"thumbs_up":4,"comments":6,"severity":"critical","severity_reasons":["title:data-loss/security"],"topic":"self-host/headless","classes":["self-host/headless","auth/session","mcp/tools","always-on/daemon"],"cosmetic_ui":false,"summary":"GitHub Copilot desktop app 1.1.22 (commit c3daa01), macOS arm64, bundled CLI 1.0.84-5 (server mode, stdio, Rust JSON-RPC engine) Single github.com account signed in to the app; gh auth status (bundled gh 2.98.0, using the app-exported…","versions":["1.1.22","1.0.84-5","2.98.0","1.0.86"],"error_snippets":["MCP tool catalog changed before tool \"github-mcp-server-web_search\" could be invoked: Failed to verify cached MCP tools: MCP request failed: Transport send error: unexpected server response: dynamic header provider faile","Execution failed: Failed to refresh MCP tools for server github-mcp-server; the last catalog is stale","[WARNING] [rust:remote::api_session_exporter] remote session batch flush failed {\"error\":\"GitHub credential authority error: GitHub token provider registration \\\"<per-process uuid>\\\" is unavailable\", …}","2.98.0, using the app-exported","session.end reason: error"],"workaround":null,"fixed_in":null,"score":98.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"anthropics/claude-code#96942","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":96942,"title":"[BUG] Windows: main process survives beforeQuitForUpdate; relaunch blocked by single-instance lock (update deadlock, reproduced on two versions)","url":"https://github.com/anthropics/claude-code/issues/96942","state":"open","state_reason":null,"labels":["bug","has repro","platform:windows","area:desktop"],"created":"2026-09-25T02:27:41Z","updated":"2026-10-25T04:20:58Z","closed":null,"reactions":3,"thumbs_up":3,"comments":9,"severity":"high","severity_reasons":["title:breakage"],"topic":"install/update","classes":["install/update","crash/hang","network"],"cosmetic_ui":false,"summary":"On Windows, the Claude Desktop app's stealth auto-update deadlocks. After an idle timeout, the app tears down its window and any running Claude Code sessions and logs beforeQuitForUpdate handler fired, going down for update, but the main…","versions":["2.2553.13","2.7032.0","2.9939.2","10.0.19045"],"error_snippets":["2026-09-24 21:43:54 [info] [stealth-update] Triggering stealth update after idle timeout","2026-09-24 21:53:54 [info] [stealth-update] Triggering stealth update after idle timeout","2026-09-24 22:13:29 [info] Not main instance, returning early from app ready","2026-09-24 22:13:36 [info] Not main instance, returning early from app ready","2026-09-24 22:13:39 [info] Not main instance, returning early from app ready"],"workaround":{"text":"`tasklist /apps | findstr /i claude ; Stop-Process -Id -Force` Then relaunch.","source":"issue body","url":"https://github.com/anthropics/claude-code/issues/96942"},"fixed_in":null,"score":97.4,"first_seen":"2026-10-22T09:27:06Z"},{"id":"google-gemini/gemini-cli#26116","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":26116,"title":"gemini stuck in thinking loop for hours","url":"https://github.com/google-gemini/gemini-cli/issues/26116","state":"open","state_reason":null,"labels":["priority/p1","area/agent","status/bot-triaged","kind/bug"],"created":"2026-04-28T14:39:00Z","updated":"2026-10-22T23:51:30Z","closed":null,"reactions":5,"thumbs_up":5,"comments":15,"severity":"high","severity_reasons":["label:p1"],"topic":"crash/hang","classes":["crash/hang","auth/session"],"cosmetic_ui":false,"summary":"its cli is thinking and thinking and loop donot breaks event it took 2 hours for completing a task which was just be done in 2 minutes in previous version, its started happening in this latest version. it also sometimes in thinking loop…","versions":["0.39.1","22.20.0","292.0"],"error_snippets":["its cli is thinking and thinking and loop donot breaks event it took 2 hours for completing a task which was just be done in 2 minutes in previous version, its started happening in this latest version. it also sometimes "],"workaround":null,"fixed_in":null,"score":96.8,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openclaw/openclaw#162768","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":162768,"title":"[Bug]: OpenAI Codex model discovery hardcodes `client_version=0.158.0`, ignoring the configured external Codex app-server","url":"https://github.com/openclaw/openclaw/issues/162768","state":"open","state_reason":null,"labels":["stale","P2","clawsweeper:no-new-fix-pr","clawsweeper:needs-maintainer-review","impact:auth-provider","issue-rating: 🦪 silver shellfish"],"created":"2026-10-01T14:37:40Z","updated":"2026-10-24T21:43:10Z","closed":null,"reactions":2,"thumbs_up":1,"comments":8,"severity":"medium","severity_reasons":["class:self-host/headless","class:auth/session"],"topic":"self-host/headless","classes":["self-host/headless","auth/session","install/update","mcp/tools"],"cosmetic_ui":false,"summary":"extensions/openai/openai-provider.ts always queries the Codex models endpoint with the managed runtime's version: However, the codex plugin officially supports external app-server binaries via plugins.entries.codex.config.appServer.command…","versions":["0.158.0","0.149.0","2026.9.7","0.159.3"],"error_snippets":["// extensions/openai/openai-provider.ts:102-105 (main)","// Keep synchronized with extensions/codex's exact @openai/codex dependency;","\"plugins\": { \"entries\": { \"codex\": { \"config\": { \"appServer\": {","extensions/openai/openai-provider.ts","plugins.entries.codex.config.appServer.command"],"workaround":null,"fixed_in":null,"score":96.3,"first_seen":"2026-10-22T09:27:06Z"},{"id":"anomalyco/opencode#56762","harness":"opencode","harness_name":"OpenCode","repo":"anomalyco/opencode","number":56762,"title":"Forbidden, OpenCode Free Model Gateway Detect Abuse Traffic","url":"https://github.com/anomalyco/opencode/issues/56762","state":"closed","state_reason":"not_planned","labels":["pending close","triaging"],"created":"2026-10-18T10:54:44Z","updated":"2026-10-20T05:51:58Z","closed":"2026-10-19T10:55:52Z","reactions":18,"thumbs_up":18,"comments":30,"severity":"medium","severity_reasons":["engagement 48","class:always-on/daemon"],"topic":"always-on/daemon","classes":["always-on/daemon"],"cosmetic_ui":false,"summary":"Forbidden, OpenCode Free Model Gateway Detect Abuse Traffic","versions":[],"error_snippets":[],"workaround":null,"fixed_in":null,"score":95.6,"first_seen":"2026-10-19T18:38:33Z"},{"id":"NousResearch/hermes-agent#135217","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":135217,"title":"[Bug]: v0.21.6 shows the v0.21.5 release date (2026.9.24) in `hermes --version` and the banner","url":"https://github.com/NousResearch/hermes-agent/issues/135217","state":"open","state_reason":null,"labels":["type/bug","comp/cli","P2","area/install-update"],"created":"2026-10-08T17:01:19Z","updated":"2026-10-25T16:56:48Z","closed":null,"reactions":2,"thumbs_up":2,"comments":9,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update","always-on/daemon","self-host/headless"],"cosmetic_ui":false,"summary":"On v0.21.6 the version line still carries the release date of v0.21.5: The version is right (it comes from the install stamp), but the date in parentheses is hermescli.releasedate, which is still \"2026.9.24\" in the v0.21.6 tag and on main…","versions":["0.21.6","0.21.5","2026.9.24","2026.10.8"],"error_snippets":["elevenlabs not set","--- errors.log (last 100 lines) ---","(file empty)","(file not found)","(not on this host: written by Hermes Desktop on the machine running the app, not by this backend. If the desktop connects to a remote/docker/SSH backend, collect it on that client machine — expected at /opt/data/logs/des"],"workaround":null,"fixed_in":null,"score":95.4,"first_seen":"2026-10-26T10:12:02Z"},{"id":"google-gemini/gemini-cli#28018","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":28018,"title":"Gemini CLI latest version is going in infinite loop for authentication","url":"https://github.com/google-gemini/gemini-cli/issues/28018","state":"open","state_reason":null,"labels":["priority/p1","area/security","status/bot-triaged","kind/bug"],"created":"2026-06-18T20:36:31Z","updated":"2026-10-02T02:24:18Z","closed":null,"reactions":7,"thumbs_up":7,"comments":7,"severity":"critical","severity_reasons":["label:security"],"topic":"crash/hang","classes":["crash/hang","auth/session"],"cosmetic_ui":false,"summary":"Whenever I'm signing in with Google - the Gemini CLI is again asking for authentication and this is never ending One time authentication only Client Information Run gemini to enter the interactive CLI, then run the /about command.","versions":["0.47.0"],"error_snippets":["to enter the interactive CLI, then run the","Whenever I'm signing in with Google - the Gemini CLI is again asking for authentication and this is never ending","Run `gemini` to enter the interactive CLI, then run the `/about` command."],"workaround":null,"fixed_in":null,"score":95.4,"first_seen":"2026-10-19T18:38:33Z"},{"id":"anthropics/claude-code#98929","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":98929,"title":"[BUG] Windows: security-guidance's sg-python.sh launches `py -3` twice per hook; each launch of the Python Install Manager alias starts an AppX update that leaks ~1.6 MB in AppXSvc, exhausting system commit (21.8 GB in 12 h)","url":"https://github.com/anthropics/claude-code/issues/98929","state":"open","state_reason":null,"labels":["bug","has repro","platform:windows","area:hooks","area:plugins"],"created":"2026-10-02T10:04:13Z","updated":"2026-10-23T18:19:14Z","closed":null,"reactions":4,"thumbs_up":4,"comments":1,"severity":"medium","severity_reasons":["class:install/update","class:crash/hang"],"topic":"install/update","classes":["install/update","crash/hang","mcp/tools","always-on/daemon"],"cosmetic_ui":false,"summary":"The security-guidance plugin (2.0.8) runs hooks/sg-python.sh on SessionStart, every UserPromptSubmit, every PostToolUse on Edit|Write|MultiEdit|NotebookEdit, PostToolUse on Bash, every Stop and every SubagentStop. On every one of those…","versions":["1.6","21.8","2.0.8","21.84"],"error_snippets":["# Microsoft-Windows-AppXDeploymentServer/Operational, event 603, once per py.exe launch, in pairs:","Edit|Write|MultiEdit|NotebookEdit","OutOfMemoryException","What makes it a Claude Code problem is the rate. With five sessions open, the hooks launched `py` often enough for **5,980 updates between 18:26 and midnight**. After 12 hours of uptime:","The update events arrive in pairs within the same second, one pair per hook: the probe launch, then the run launch."],"workaround":null,"fixed_in":null,"score":95.0,"first_seen":"2026-10-25T09:11:37Z"},{"id":"anomalyco/opencode#48645","harness":"opencode","harness_name":"OpenCode","repo":"anomalyco/opencode","number":48645,"title":"Regression in 1.18.30: every prompt crashes with TypeError in SystemPrompt.environment (\"a.name\") — 1.18.18 works fine","url":"https://github.com/anomalyco/opencode/issues/48645","state":"closed","state_reason":"completed","labels":[],"created":"2026-09-12T10:53:51Z","updated":"2026-09-21T21:41:31Z","closed":"2026-09-21T21:41:31Z","reactions":18,"thumbs_up":18,"comments":10,"severity":"high","severity_reasons":["title:breakage"],"topic":"install/update","classes":["install/update","crash/hang","self-host/headless","mcp/tools"],"cosmetic_ui":false,"summary":"After updating to opencode 1.18.30 via pacman, every single prompt fails immediately with: Failed to send prompt. Unexpected server error. Check server logs for details. This happens on a brand new session, on the very first message, with…","versions":["1.18.30","1.18.18","1.18.7","1.18.18-1"],"error_snippets":["SystemPrompt.environment","After updating to opencode 1.18.30 via pacman, every single prompt fails immediately with:","Failed to send prompt. Unexpected server error. Check server logs for details.","TypeError: undefined is not an object (evaluating 'a.name')","at SystemPrompt.environment (/$bunfs/root/chunk-5gp4k9hv.js:50:13096)"],"workaround":{"text":"Downgrading to 1.18.18 (via the cached pacman package) fixes it immediately — same machine, same everything else.","source":"issue body","url":"https://github.com/anomalyco/opencode/issues/48645"},"fixed_in":null,"score":94.7,"first_seen":"2026-10-19T18:38:33Z"},{"id":"anomalyco/opencode#56297","harness":"opencode","harness_name":"OpenCode","repo":"anomalyco/opencode","number":56297,"title":"Unable to install OpenCode on windows with winget command","url":"https://github.com/anomalyco/opencode/issues/56297","state":"open","state_reason":null,"labels":["reproduced"],"created":"2026-10-17T03:28:42Z","updated":"2026-10-22T06:03:54Z","closed":null,"reactions":6,"thumbs_up":6,"comments":2,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update","crash/hang"],"cosmetic_ui":false,"summary":"Unable to install OpenCode on windows with winget command I follow the instruction described in Changelog: https://github.com/anomalyco/opencode/blob/dev/CHANGELOG.md#highlights-3 The root cause…","versions":[],"error_snippets":["Unable to install OpenCode on windows with winget command"],"workaround":null,"fixed_in":null,"score":93.4,"first_seen":"2026-10-25T09:11:37Z"},{"id":"openclaw/openclaw#145252","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":145252,"title":"[Tracking] 2026.9.3 / 2026.9.4 update, upgrade and recovery reliability","url":"https://github.com/openclaw/openclaw/issues/145252","state":"open","state_reason":null,"labels":["maintainer","clawsweeper:needs-info","impact:crash-loop","P0","issue-rating: 🦐 gold shrimp","impact:ux-release-blocker","clawtributor-review"],"created":"2026-09-11T20:41:52Z","updated":"2026-10-22T01:34:40Z","closed":null,"reactions":2,"thumbs_up":0,"comments":13,"severity":"critical","severity_reasons":["label:p0"],"topic":"install/update","classes":["install/update","crash/hang","network"],"cosmetic_ui":false,"summary":"Track update, upgrade, Doctor, migration, rollback and restart reliability around OpenClaw 2026.9.3 / 2026.9.4. This coordination index does not replace the linked reports or their evidence, and does not authorize closing, merging or…","versions":["2026.9.3","2026.9.4"],"error_snippets":["Track update, upgrade, Doctor, migration, rollback and restart reliability around OpenClaw 2026.9.3 / 2026.9.4. This coordination index does not replace the linked reports or their evidence, and does not authorize closin","Add coordination notes outside the managed section below. Automation preserves that text. The delimited index is managed; conflicting edits inside it stop synchronization for review.","Open a linked item for its full title and discussion. Every reference, current priority, assignee and scope is retained below.","Maintenance: configured to synchronize with the hourly upgrade-dashboard refresh. Unchanged snapshots do not create edits; failed reads retain the previous published snapshot.","This is the known dashboard-linked scope, including related candidates and contextual references—not an exhaustive repository-wide search. A shared label or symptom does not prove duplication. An open PR is not a complet"],"workaround":null,"fixed_in":null,"score":93.3,"first_seen":"2026-10-18T04:33:05Z"},{"id":"NousResearch/hermes-agent#32903","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":32903,"title":"openai-codex provider crashes: SDK parse_response fails on null output from Codex backend","url":"https://github.com/NousResearch/hermes-agent/issues/32903","state":"closed","state_reason":"completed","labels":["type/bug","comp/agent","provider/openai","P3"],"created":"2026-05-27T00:49:26Z","updated":"2026-10-17T13:06:19Z","closed":"2026-05-27T03:06:26Z","reactions":25,"thumbs_up":25,"comments":17,"severity":"high","severity_reasons":["title:breakage"],"topic":"crash/hang","classes":["crash/hang","install/update","auth/session"],"cosmetic_ui":false,"summary":"Hermes v0.14.0 (latest, updated 2026-05-27 with 471 commits pulled) crashes on every hermes chat call when using openai-codex provider with gpt-5.5. The ChatGPT Codex backend (chatgpt.com/backend-api/codex) returns output: null in…","versions":["0.14.0","2.38.0","2026.5.16","2.24.0"],"error_snippets":["Error: 'NoneType' object is not iterable","output: null","Hermes v0.14.0 (latest, updated 2026-05-27 with 471 commits pulled) crashes on every `hermes chat` call when using `openai-codex` provider with `gpt-5.5`.","The ChatGPT Codex backend (`chatgpt.com/backend-api/codex`) returns `output: null` in `response.completed` SSE stream events. The OpenAI Python SDK's `parse_response()` at `openai/lib/_parsing/_responses.py:61` does `for"],"workaround":null,"fixed_in":null,"score":93.2,"first_seen":"2026-10-18T04:29:32Z"},{"id":"anomalyco/opencode#49827","harness":"opencode","harness_name":"OpenCode","repo":"anomalyco/opencode","number":49827,"title":"migrate-v1: non-git sessions in \"global\" project hidden after V1→V2 upgrade","url":"https://github.com/anomalyco/opencode/issues/49827","state":"open","state_reason":"reopened","labels":[],"created":"2026-09-18T17:59:43Z","updated":"2026-10-23T06:33:57Z","closed":null,"reactions":3,"thumbs_up":3,"comments":17,"severity":"medium","severity_reasons":["engagement 20","class:install/update"],"topic":"install/update","classes":["install/update","self-host/headless"],"cosmetic_ui":false,"summary":"After upgrading from V1.18.31 to V2.0.8, sessions that V1 stored in the global project (sessions created in non-git directories) become invisible in the TUI session picker, because V2 creates a per-directory project and the picker filters…","versions":["2.0.8","1.18.31"],"error_snippets":["(\"List all sessions\") shows everything, including the"],"workaround":{"text":"Before repair: 104 sessions total, 93 with `project_id = 'global'` (81 in `/home/wjf`, 8 in `/tmp/opencode`, 3 in `/home/wjf/novel`, 1 in `/home/wjf/videos`). - Sessions in a git repository were unaffected: `/home/wjf/fadebreak` kept a stable project id across the upgrade and stayed visible. -…","source":"issue body","url":"https://github.com/anomalyco/opencode/issues/49827"},"fixed_in":null,"score":93.2,"first_seen":"2026-10-23T09:23:24Z"},{"id":"anthropics/claude-code#94758","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":94758,"title":"Claude Desktop 3p + custom gateway (`ANTHROPIC_BASE_URL`): sandbox network pinned to gateway + api.anthropic.com, `sandbox.network.allowedDomains` ignored even after full app restart","url":"https://github.com/anthropics/claude-code/issues/94758","state":"open","state_reason":null,"labels":["bug","has repro","platform:macos","area:networking","area:desktop","area:sandbox"],"created":"2026-09-16T12:53:52Z","updated":"2026-09-16T12:55:41Z","closed":null,"reactions":12,"thumbs_up":12,"comments":0,"severity":"medium","severity_reasons":["class:always-on/daemon"],"topic":"always-on/daemon","classes":["always-on/daemon","install/update","auth/session","network"],"cosmetic_ui":false,"summary":"Claude Desktop (enterprise 3p deployment, deploymentMode: \"3p\"), macOS (Apple Silicon), app 1.24012.1 Embedded Claude Code: 2.1.236 (second machine: 2.1.270), claude-desktop-3p entrypoint Inference: custom gateway via…","versions":["2.1.236","2.1.270"],"error_snippets":["deny network-outbound <host>:443 (user denied)","curl: (56) CONNECT tunnel failed, response 403","Sandboxed Bash egress is pinned to exactly two hosts. From a sandboxed Bash command:","Blocked hosts fail silently at the srt proxy (no permission prompt appears; on another machine the same blocks surfaced as `deny network-outbound :443 (user denied)` and `curl: (56) CONNECT tunnel failed, response 403`).","4. Re-run the probes from sandboxed Bash → `api.openai.com` and `pypi.org` still blocked (000/exit 56)."],"workaround":null,"fixed_in":null,"score":93.0,"first_seen":"2026-10-18T04:33:44Z"},{"id":"NousResearch/hermes-agent#132817","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":132817,"title":"Pain cluster: one transient 429 / entitlement cooldown benches a credential for days — no reset-time probe, no visibility, no pointer to `hermes auth reset` (13 reports this week)","url":"https://github.com/NousResearch/hermes-agent/issues/132817","state":"open","state_reason":null,"labels":["type/bug","comp/agent","comp/cli","provider/openai","area/auth","P2","sweeper:risk-security-boundary","area/profiles","from-pain-miner"],"created":"2026-10-04T16:11:23Z","updated":"2026-10-20T07:46:01Z","closed":null,"reactions":2,"thumbs_up":2,"comments":8,"severity":"critical","severity_reasons":["title:data-loss/security"],"topic":"auth/session","classes":["auth/session","network","always-on/daemon","crash/hang"],"cosmetic_ui":false,"summary":"A single transient 429 / modelentitlement cooldown on one credential benches that credential (and often the whole provider row, or every auxiliary call) for hours to days, with no reset-time probe, no visible \"in cooldown until T because…","versions":[],"error_snippets":["s, or a fallback model that never lets go, and then hand-edit","HTTP 429: The usage limit has been reached","last_error_*","/ reset headers everywhere (pool, aux, kanban); at","This is a pain-miner umbrella: the individual mechanisms are filed below. What is not filed is that this is **one class** (cooldown persistence + scope + visibility + reset), that `hermes auth reset ` already exists (cre"],"workaround":null,"fixed_in":null,"score":91.6,"first_seen":"2026-10-18T04:32:26Z"},{"id":"aaif-goose/goose#12703","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":12703,"title":"Render A2UI embedded resources from MCP tool results in goose desktop","url":"https://github.com/aaif-goose/goose/issues/12703","state":"open","state_reason":null,"labels":[],"created":"2026-10-06T01:30:18Z","updated":"2026-10-22T22:04:35Z","closed":null,"reactions":0,"thumbs_up":0,"comments":15,"severity":"medium","severity_reasons":["engagement 15","class:mcp/tools"],"topic":"mcp/tools","classes":["mcp/tools","self-host/headless","install/update","crash/hang"],"cosmetic_ui":false,"summary":"Proposal: goose doesn't support A2UI today. This issue proposes adding it to goose desktop, starting with rendering A2UI surfaces returned by MCP servers. The first version would be off by default and limited to the A2UI basic catalog…","versions":[],"error_snippets":["capabilities.extensions","call (name, surfaceId, sourceComponentId, context) would go through the existing","path to the extension that produced the surface. That path requires a tool visible to apps (the default when","ExtensionManager::dispatch_tool_call","A2UI would render in goose's own window, not a sandbox, so goose would treat every payload as untrusted data:"],"workaround":null,"fixed_in":null,"score":91.5,"first_seen":"2026-10-26T10:12:02Z"},{"id":"omacom/omarchy#11590","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":11590,"title":"gnome-keyring-daemon crashes with NULL client pointer in secret service","url":"https://github.com/omacom/omarchy/issues/11590","state":"open","state_reason":null,"labels":[],"created":"2026-09-13T00:06:45Z","updated":"2026-10-22T01:44:00Z","closed":null,"reactions":2,"thumbs_up":2,"comments":11,"severity":"high","severity_reasons":["title:breakage"],"topic":"always-on/daemon","classes":["always-on/daemon","crash/hang","network"],"cosmetic_ui":false,"summary":"gnome-keyring-daemon crashed with SIGABRT on 2026-09-13 at 01:38:02 CEST while handling D-Bus requests. Signal: SIGABRT (from glogdefaulthandler calling abort()) Root cause: NULL client pointer in gkdsecretservicegetpkcs11session cascading…","versions":["50.0-1"],"error_snippets":["gkd_secret_service_get_pkcs11_session: assertion 'client' failed","gkd_secret_service_publish_dispatch: assertion 'client' failed","g_variant_new: expected GVariant of type 'v' but received value has type '(null)'","This is likely a gnome-keyring upstream bug (NULL pointer validation), but reporting here as well in case Omarchy's packaging or D-Bus integration is involved."],"workaround":null,"fixed_in":null,"score":91.4,"first_seen":"2026-10-19T18:38:33Z"},{"id":"NousResearch/hermes-agent#125437","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":125437,"title":"Pain cluster: a failed update leaves a half-applied install with a raw error and no in-product recovery — 5 mechanisms, 15 Discord threads this week, every fix is a hand-typed recipe","url":"https://github.com/NousResearch/hermes-agent/issues/125437","state":"open","state_reason":null,"labels":["type/bug","comp/cli","P1","sweeper:risk-compatibility","comp/desktop","area/install-update","from-pain-miner","os/windows"],"created":"2026-09-27T16:09:32Z","updated":"2026-10-25T16:10:12Z","closed":null,"reactions":1,"thumbs_up":1,"comments":16,"severity":"high","severity_reasons":["label:p1"],"topic":"install/update","classes":["install/update","always-on/daemon"],"cosmetic_ui":false,"summary":"When hermes update / the Desktop updater fails, the install is left half-applied (Desktop launches then closes, gateway venv missing pydanticcore, git checkout carrying partial-clone config) and the product prints the raw underlying error…","versions":[],"error_snippets":["✗ Failed to fetch updates from origin. BUG: builtin/pack-objects.c:4842: should_include_obj …","tar.exe: Error opening archive: Can't initialize filter; unable to run program \"bzip2 -d\"","[X] failed to extract pinned git archive","index-pack failed","This is a pain-miner umbrella: the individual mechanisms are already filed (linked below). What is not filed is that the *class* has no in-product self-recovery and that the fixes for every mechanism are sitting in open "],"workaround":null,"fixed_in":null,"score":91.3,"first_seen":"2026-10-26T10:12:02Z"},{"id":"NousResearch/hermes-agent#136567","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":136567,"title":"Telegram DM topics: opening message of a newly created topic is delivered thread-less → lobby recovery misroutes it to the previous topic","url":"https://github.com/NousResearch/hermes-agent/issues/136567","state":"closed","state_reason":"completed","labels":["type/bug","comp/gateway","platform/telegram","P2","sweeper:risk-session-state","sweeper:risk-message-delivery","area/sessions","cause/upstream"],"created":"2026-10-11T11:11:59Z","updated":"2026-10-20T04:48:46Z","closed":"2026-10-20T04:47:55Z","reactions":8,"thumbs_up":8,"comments":32,"severity":"medium","severity_reasons":["engagement 40","class:always-on/daemon"],"topic":"always-on/daemon","classes":["always-on/daemon","install/update","network"],"cosmetic_ui":false,"summary":"In a forum-enabled bot DM (Telegram \"Threads\" feature), when the user creates a new topic via the client's \"New thread\" composer, Telegram now delivers the opening message without messagethreadid (thread-less), while the forumtopiccreated…","versions":["0.21.6+607.gc573316"],"error_snippets":["event.source.thread_id","In a forum-enabled bot DM (Telegram \"Threads\" feature), when the user creates a new topic via the client's \"New thread\" composer, Telegram now delivers the **opening message without `message_thread_id`** (thread-less), w","Hermes' lobby recovery (`_recover_telegram_topic_thread_id`, `gateway/run_topics.py`) then pins the thread-less message to the user's **most-recent bound topic**: the message is processed in the *previous* session (inter","User-visible symptom: \"when I write from 'New thread' it keeps answering in the last active thread and creates an empty section.\"","A local user plugin implements exactly this (narrow PTB `filters.MessageFilter` for `forum_topic_created` via `register_platform_handler`, plus a `pre_gateway_dispatch` hook that rewrites `event.source.thread_id`); verif"],"workaround":null,"fixed_in":null,"score":89.7,"first_seen":"2026-10-18T04:29:32Z"},{"id":"NousResearch/hermes-agent#72298","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":72298,"title":"[Bug]: Hermes shows passwords in Telegram chat","url":"https://github.com/NousResearch/hermes-agent/issues/72298","state":"open","state_reason":null,"labels":["type/bug","comp/agent","comp/gateway","tool/browser","platform/telegram","P2","needs-repro","sweeper:risk-message-delivery","sweeper:risk-security-boundary"],"created":"2026-07-26T23:29:05Z","updated":"2026-09-17T13:27:22Z","closed":null,"reactions":11,"thumbs_up":8,"comments":5,"severity":"medium","severity_reasons":["engagement 16","class:always-on/daemon"],"topic":"always-on/daemon","classes":["always-on/daemon","auth/session"],"cosmetic_ui":false,"summary":"Hi! I'm developing a skill that uses the following stack: Kimi K3; from OpenCode Go Bitwarden; for secrets (using the \"official\" resource from Hermes docs) Browserbase headless; integrated in Hermes (free) without API key - as is. Hermes…","versions":["0.19.0","2026.7.20","3.10.12"],"error_snippets":[],"workaround":null,"fixed_in":null,"score":88.8,"first_seen":"2026-10-18T04:29:32Z"},{"id":"NousResearch/hermes-agent#138649","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":138649,"title":"[Bug]: Telegram init hangs pre-socket inside gateway (no SYN, healthy loop, standalone 0.3s) — recurring #63309-family variant","url":"https://github.com/NousResearch/hermes-agent/issues/138649","state":"open","state_reason":null,"labels":["type/bug","comp/gateway","comp/plugins","platform/telegram","P1","sweeper:risk-message-delivery"],"created":"2026-10-14T22:05:44Z","updated":"2026-10-15T11:48:36Z","closed":null,"reactions":1,"thumbs_up":0,"comments":5,"severity":"high","severity_reasons":["label:p1","title:breakage"],"topic":"always-on/daemon","classes":["always-on/daemon","crash/hang","mcp/tools","network","install/update","auth/session"],"cosmetic_ui":false,"summary":"Telegram init hangs pre-socket inside the gateway process after a routine network blip, while the identical stack (same venv, same token, same transports, same SSL context) connects in 0.3s standalone. Appears to be a new variant of the…","versions":[],"error_snippets":["ConnectError('')","TimeoutError","executor threads idle in","primitive bound to a different (earlier) loop/event-loop context — the classic \"works standalone, hangs in gateway\" signature: no fd, no timer, loop idle in","tasks after the first fatal error ("],"workaround":null,"fixed_in":null,"score":88.3,"first_seen":"2026-10-18T04:32:26Z"},{"id":"openclaw/openclaw#160548","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":160548,"title":"[Bug]: 2026.9.6 prepared-model-catalog worker leaks ~1 GiB per 5 min; each memory reclamation supersedes the runtime publication and kills every waiting turn","url":"https://github.com/openclaw/openclaw/issues/160548","state":"open","state_reason":null,"labels":["stale","impact:message-loss","impact:crash-loop","P0","issue-rating: 🦪 silver shellfish"],"created":"2026-09-28T16:26:12Z","updated":"2026-10-22T03:44:59Z","closed":null,"reactions":2,"thumbs_up":1,"comments":14,"severity":"critical","severity_reasons":["label:p0"],"topic":"crash/hang","classes":["crash/hang","always-on/daemon","install/update"],"cosmetic_ui":false,"summary":"Since upgrading to 2026.9.6 (eb377ac) the prepared-model-catalog.worker.js thread grows by roughly 1 GiB every 5 minutes with no user action, reaches 8 to 13 GiB, and the gateway's memory-pressure reclamation then resets it to under 2 GiB…","versions":["2026.9.6","8.59","7.62","11.7"],"error_snippets":["... same sawtooth every 45 to 60 min through the day ...","11:01:56.587 memory pressure: level=warning reason=rss_threshold rss=5.82 GiB heap=565.9 MiB threshold=4.55 GiB workerCount=9","11:03:23.281 {'reclamationKind': 'maintenance-plan', 'workerThreadId': 76, 'elapsedMs': 1436, 'outcome': 'resolved'}","11:03:26.030 message dispatch completed: channel=webchat sessionKey=agent:main:... source=replyResolver outcome=error duration=47453ms","error=\"Error: prepared model runtime publication was superseded for <home>/.openclaw/agents/main/agent\""],"workaround":null,"fixed_in":null,"score":88.0,"first_seen":"2026-10-24T08:54:58Z"},{"id":"anthropics/claude-code#98310","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":98310,"title":"Remote Control: unarchived server-mode sessions are never re-dispatched to the running host — messages hang at \"Sending…\", then \"Claude Code on <machine> is offline\"","url":"https://github.com/anthropics/claude-code/issues/98310","state":"open","state_reason":null,"labels":["bug","has repro","platform:linux"],"created":"2026-09-30T04:10:55Z","updated":"2026-10-25T11:08:37Z","closed":null,"reactions":3,"thumbs_up":3,"comments":3,"severity":"high","severity_reasons":["title:breakage"],"topic":"self-host/headless","classes":["self-host/headless","crash/hang","always-on/daemon","network"],"cosmetic_ui":false,"summary":"A session served by a long-running claude remote-control host (server mode, same-dir spawn) can't be used again after it's archived and then unarchived from claude.ai. The web UI accepts the unarchive, but a message sent to the session…","versions":[],"error_snippets":["POST /v1/environments/<env_id>/bridge/reconnect {\"session_id\":\"cse_…\"} → 200",", so the child exits with code 1 and the host logs","A session served by a long-running `claude remote-control` host (server mode, same-dir spawn) can't be used again after it's archived and then unarchived from claude.ai. The web UI accepts the unarchive, but a message se","2. Start a session from claude.ai and exchange a message or two.","3. Archive the session from claude.ai. The host logs `Session completed …` and the child process exits, as intended since #40598."],"workaround":{"text":"`claude remote-control --session-id ` in the same directory. This starts a separate single-session host. - Calling `bridge/reconnect` for the session, as above. This puts it back into the existing host.","source":"issue body","url":"https://github.com/anthropics/claude-code/issues/98310"},"fixed_in":null,"score":87.8,"first_seen":"2026-10-18T04:33:44Z"},{"id":"anthropics/claude-code#101369","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":101369,"title":"[BUG][REGRESSION] Broken Projects upgrade wiped out classic projects with no no opt-out","url":"https://github.com/anthropics/claude-code/issues/101369","state":"open","state_reason":null,"labels":["bug","area:cowork","platform:web"],"created":"2026-10-11T21:51:41Z","updated":"2026-10-24T12:43:31Z","closed":null,"reactions":4,"thumbs_up":4,"comments":5,"severity":"critical","severity_reasons":["title:data-loss/security"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"This is not a cosmetic redesign. My classic claude.ai chat projects, which I used as sealed folders of separate private conversations with their own instructions, were converted to the coordinator-based Projects without my consent, without…","versions":[],"error_snippets":["1. Old project chats now also appear in the main sidebar, mixed in with standalone chats, with no project label. You can't group or filter by project.","3. The coordinator can see every thread in the project. Chats that used to be separate are no longer private from it.","Anthropic's announcement and Help Center promise that existing chat projects \"keep working as they do today\". That promise is meaningless once the upgrade lands: the classic behaviour is gone entirely, nothing explains w","5. Open one of the old chats from the Threads panel. It opens read-only in a side pane, ending with \"End of original chat · Reply to continue\", with a box labelled \"Reply to start a new thread…\".","7. Earlier in the same project, the coordinator confirmed: \"I can see what's written in every thread, even when I stay out of it, so threads here aren't private from me the way your old project chats were\", and \"Every ne"],"workaround":null,"fixed_in":null,"score":87.6,"first_seen":"2026-10-19T18:38:33Z"},{"id":"kirodotdev/Kiro#11754","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":11754,"title":"Reliability regression: file edits fail, IDE freezes, spec operations die with network errors, and in-app bug reporter is broken","url":"https://github.com/kirodotdev/Kiro/issues/11754","state":"open","state_reason":null,"labels":["pending-maintainer-response"],"created":"2026-10-02T20:07:12Z","updated":"2026-10-07T19:00:41Z","closed":null,"reactions":0,"thumbs_up":0,"comments":15,"severity":"high","severity_reasons":["title:breakage"],"topic":"install/update","classes":["install/update","crash/hang"],"cosmetic_ui":false,"summary":"Windows 11 Business 25H2 Summary Over the past several weeks Kiro has become unreliable across core operations. File edits fail, the IDE becomes unresponsive, and spec (design/task) operations fail repeatedly with a backend network error —…","versions":["1.1.14"],"error_snippets":["Summary Over the past several weeks Kiro has become unreliable across core operations. File edits fail, the IDE becomes unresponsive, and spec (design/task) operations fail repeatedly with a backend network error — while","File edits fail — [edits don't apply / hang silently / throw an error: paste exact text if any].","Spec operations fail with a network error — generating a design document from existing requirements failed three consecutive times with Client network error calling Q. Each attempt re-ran ~25+ tool calls (re-reading requ","In-app bug reporter is broken — the report page errors out, so bugs can't be filed through the intended channel.","Observe: full re-exploration of the codebase runs, then the operation fails with Client network error calling Q."],"workaround":null,"fixed_in":null,"score":87.6,"first_seen":"2026-10-19T18:38:33Z"},{"id":"google-gemini/gemini-cli#28598","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":28598,"title":"CLI crashes on startup when ran inside a Git repository and MacOS seatbelt is enabled","url":"https://github.com/google-gemini/gemini-cli/issues/28598","state":"open","state_reason":null,"labels":["priority/p1","area/platform","status/bot-triaged","kind/bug","effort/medium"],"created":"2026-07-30T11:28:54Z","updated":"2026-10-24T14:17:29Z","closed":null,"reactions":3,"thumbs_up":3,"comments":12,"severity":"high","severity_reasons":["label:p1","title:breakage"],"topic":"crash/hang","classes":["crash/hang","network"],"cosmetic_ui":false,"summary":"Running Gemini CLI with MacOS seatbelt enabled: And then starting Gemini CLI in a folder that contains a Git repository instantly crashes the CLI. Folder without a Git repository: Initialising this folder with a Git repo: Not to crash…","versions":["0.53.0"],"error_snippets":["│ Sandbox sandbox-exec (permissive-open) │","to enter the interactive CLI, then run the","Running Gemini CLI with MacOS seatbelt enabled:","Run `gemini` to enter the interactive CLI, then run the `/about` command.","`GOOGLE_CLOUD_PROJECT` Environment variable is set."],"workaround":null,"fixed_in":null,"score":87.3,"first_seen":"2026-10-19T18:38:33Z"},{"id":"NousResearch/hermes-agent#131055","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":131055,"title":"Linux Desktop: second-instance launches poison the sandbox fallback marker → sticky --no-sandbox → renderer SIGILL loop (/dev/shm ESRCH)","url":"https://github.com/NousResearch/hermes-agent/issues/131055","state":"open","state_reason":null,"labels":["type/bug","P2","sweeper:risk-security-boundary","comp/desktop","desktop/shell"],"created":"2026-10-02T00:55:55Z","updated":"2026-10-24T20:23:39Z","closed":null,"reactions":1,"thumbs_up":1,"comments":23,"severity":"medium","severity_reasons":["engagement 24"],"topic":"crash/hang","classes":["crash/hang"],"cosmetic_ui":false,"summary":"On Linux, launching Hermes Desktop again while it is already running (the launcher, gtk-launch, or a hermes:// deeplink that starts the binary) can leave windows-sandbox-fallback.json stuck at {\"state\":\"fallback\",\"reason\":\"boot-loop\"}…","versions":[],"error_snippets":["ERROR:base/memory/platform_shared_memory_region_posix.cc:214] Creating shared memory in /dev/shm/.org.chromium.Chromium.XXXXXX failed: No such process (3)","ERROR:base/memory/platform_shared_memory_region_posix.cc:217] Unable to access(W_OK|X_OK) /dev/shm: No such process (3)","FATAL:base/memory/platform_shared_memory_region_posix.cc:219] This is frequently caused by incorrect permissions on /dev/shm. Try 'sudo chmod 1777 /dev/shm' to fix.","[deeplink] delivery failed: Object has been destroyed","[hermes] sandbox fallback enabled (sticky-fallback); launching with --no-sandbox (#38216, #121954)"],"workaround":null,"fixed_in":null,"score":85.4,"first_seen":"2026-10-18T04:32:26Z"},{"id":"openclaw/openclaw#48920","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":48920,"title":"[Bug]: Live Docs are ahead of release","url":"https://github.com/openclaw/openclaw/issues/48920","state":"open","state_reason":null,"labels":["bug","regression","clawsweeper:no-new-fix-pr","clawsweeper:needs-product-decision","clawsweeper:source-repro","P0","issue-rating: 🦞 diamond lobster","impact:ux-release-blocker"],"created":"2026-03-17T10:14:14Z","updated":"2026-10-23T05:17:04Z","closed":null,"reactions":6,"thumbs_up":4,"comments":11,"severity":"critical","severity_reasons":["label:p0"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"Regression (worked before, now fails) Heartbeat IsolatedSessions is in the live docs but not in the latest version 2026.3.13 Add IsolatedSessions to your heartbeat config as per the docs…","versions":["2026.3.13"],"error_snippets":["You will get `agents.defaults.heartbeat: Unrecognized key: \"isolatedSession\"` errored,"],"workaround":null,"fixed_in":null,"score":85.0,"first_seen":"2026-10-18T04:30:03Z"},{"id":"omacom/omarchy#14469","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":14469,"title":"Qt 6.12.0 update breaks the shell: quickshell 0.3.1-1 still built against Qt 6.11.2","url":"https://github.com/omacom/omarchy/issues/14469","state":"closed","state_reason":"completed","labels":[],"created":"2026-10-07T19:41:59Z","updated":"2026-10-08T17:13:48Z","closed":"2026-10-08T17:13:48Z","reactions":16,"thumbs_up":13,"comments":16,"severity":"medium","severity_reasons":["engagement 32","class:install/update"],"topic":"install/update","classes":["install/update","always-on/daemon","mcp/tools"],"cosmetic_ui":false,"summary":"AMD Ryzen 9 5900X, NVIDIA RTX 2080 Ti, Omarchy 4.0.4 Today's update pulls Qt 6.12.0 from extra, but quickshell 0.3.1-1 is still built against Qt 6.11.2. After the update and a shell restart: The bar is gone. Every menu (Super+Space…","versions":["6.12.0","0.3.1-1","6.11.2","4.0.4"],"error_snippets":["WARN scene: @plugins/bar/Bar.qml[81:-1]: TypeError: Cannot read property 'active' of undefined","WARN scene: @plugins/bar/Bar.qml[75:3]: Unable to assign [undefined] to QColor","WARN scene: @Ui/Button.qml[57:-1]: TypeError: Cannot read property 'background' of undefined","WARN scene: @Ui/PanelToolTip.qml[22:-1]: TypeError: Cannot read property 'background' of undefined","Today's update pulls Qt 6.12.0 from extra, but `quickshell 0.3.1-1` is still built against Qt 6.11.2. After the update and a shell restart:"],"workaround":{"text":"downgrade the qt6- packages plus fcitx5-qt, python-pyqt6, qt6ct and telegram-desktop from the pacman cache, then add them to IgnorePkg until quickshell is rebuilt.","source":"issue body","url":"https://github.com/omacom/omarchy/issues/14469"},"fixed_in":null,"score":84.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openclaw/openclaw#162119","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":162119,"title":"Codex intermittently returns 403 owner-verification error after an in-place model switch","url":"https://github.com/openclaw/openclaw/issues/162119","state":"open","state_reason":null,"labels":["P1","impact:security","impact:message-loss","impact:auth-provider","issue-rating: 🦪 silver shellfish"],"created":"2026-09-30T20:44:09Z","updated":"2026-10-10T15:01:28Z","closed":null,"reactions":2,"thumbs_up":1,"comments":14,"severity":"critical","severity_reasons":["label:security"],"topic":"auth/session","classes":["auth/session","always-on/daemon","install/update","network"],"cosmetic_ui":false,"summary":"On OpenClaw 2026.9.7 (c074824), an existing Codex-backed conversation intermittently fails before a reply with: The first failure followed an in-place session model switch from openai/gpt-6-astra to openai/gpt-6-sol. A…","versions":["2026.9.7"],"error_snippets":["unexpected status 403 Forbidden: Codex cannot verify the owner of this model request. Start a fresh request before retrying.","CodexThreadPolicyHandoffError","was added to Agent A's selectable model and allow policy. The model of the **existing** conversation was patched in place at",": the next turn failed with",": first 403 owner-verification refusal on"],"workaround":null,"fixed_in":null,"score":84.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"google-gemini/gemini-cli#29682","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":29682,"title":"Gemini CLI v0.63.0 - breaks yolo mode","url":"https://github.com/google-gemini/gemini-cli/issues/29682","state":"open","state_reason":null,"labels":["status/need-triage"],"created":"2026-10-08T14:53:01Z","updated":"2026-10-22T09:13:29Z","closed":null,"reactions":8,"thumbs_up":5,"comments":9,"severity":"medium","severity_reasons":["engagement 17"],"topic":"auth/session","classes":["auth/session"],"cosmetic_ui":false,"summary":"⚠ ⚠️ CRITICAL SECURITY WARNING: Untrusted Command Flags Detected The following flags/arguments were sourced from external untrusted input (e.g., Buganizer ticket, external document): Executing commands with flags sourced from untrusted…","versions":["0.63.0"],"error_snippets":["to enter the interactive CLI, then run the","The following flags/arguments were sourced from external untrusted input (e.g., Buganizer ticket, external document):","Executing commands with flags sourced from untrusted input can execute arbitrary code on your machine. Carefully verify these flags before approving.","I expect that in yolo mode, no confirmations will be required to continue operation. Or, at the very least, there will be a startup parameter to disable additional confirmations.","Run `gemini` to enter the interactive CLI, then run the `/about` command."],"workaround":{"text":"then rolled back to v0.62.0 via npm install -g @google/gemini-cli@v0.62 which works.","source":"comment by @crypt0g30rgy (2 reactions)","url":"https://github.com/google-gemini/gemini-cli/issues/29682#issuecomment-6074128513"},"fixed_in":null,"score":83.6,"first_seen":"2026-10-19T18:38:33Z"},{"id":"omacom/omarchy#6866","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":6866,"title":"Migration 1786643346 loops after reboot on stale Chromium SingletonLock","url":"https://github.com/omacom/omarchy/issues/6866","state":"open","state_reason":null,"labels":["bug"],"created":"2026-08-14T12:54:41Z","updated":"2026-10-08T04:17:47Z","closed":null,"reactions":9,"thumbs_up":9,"comments":4,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update","network"],"cosmetic_ui":false,"summary":"Omarchy: 4.0.0rc5-1 Kernel: 7.1.8-arch1-3 CPU: AMD Ryzen AI MAX+ PRO 395 GPU: AMD Strix Halo / Radeon 8060S Migration 1786643346 repeatedly asks me to close the browser windows and continue even after Chromium is completely stopped…","versions":[],"error_snippets":["no longer existed, and the socket target under","was rebound to the pinned Copy URL extension id","Migration `1786643346` repeatedly asks me to close the browser windows and continue even after Chromium is completely stopped. Pressing Continue just returns to the same prompt.","There was no Chromium-family process running, PID `112033` no longer existed, and the socket target under `/tmp` no longer existed.","The migration nevertheless considers the profile open because `profile_open()` tests `-L \"$profile/SingletonLock\"`. That remains true for a stale symlink, so the `while affected_profile_open` loop cannot make progress re"],"workaround":null,"fixed_in":null,"score":83.3,"first_seen":"2026-10-19T18:38:33Z"},{"id":"kirodotdev/Kiro#11980","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":11980,"title":"[CLI V3 Regression][Windows][2.31.1] `skill://` resources not recognized and absolute paths resolved incorrectly","url":"https://github.com/kirodotdev/Kiro/issues/11980","state":"open","state_reason":null,"labels":["os: windows","pending-maintainer-response","cli"],"created":"2026-10-16T17:13:39Z","updated":"2026-10-23T02:37:18Z","closed":null,"reactions":0,"thumbs_up":0,"comments":6,"severity":"high","severity_reasons":["title:breakage"],"topic":"install/update","classes":["install/update","mcp/tools","crash/hang"],"cosmetic_ui":false,"summary":"Window 10/11 Kiro CLI 2.31.1 Kiro CLI 2.31.1 V2 engine: Working V3 engine: Affected Custom agent configuration: JSON (resources) Kiro CLI V3 appears to have two regressions in custom agent resource handling compared with V2. In V2, skills…","versions":["2.31.1"],"error_snippets":["for the tested paths results in no skill entry in","In V2, skills explicitly declared in an agent's `resources` are visible through `/context show`.","For example, a resource beginning with:","2. Create or use an existing skill:","8. Compare with V2, where explicitly declared skill resources are visible."],"workaround":{"text":"Using `file://` as a workaround is not equivalent to registering a skill for progressive loading.","source":"issue body","url":"https://github.com/kirodotdev/Kiro/issues/11980"},"fixed_in":null,"score":83.0,"first_seen":"2026-10-23T09:23:24Z"},{"id":"github/copilot-cli#4998","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":4998,"title":"Copilot CLI unusable after macOS update/reboot because `.mcp-writer.binding` persists stale filesystem device ID","url":"https://github.com/github/copilot-cli/issues/4998","state":"closed","state_reason":"completed","labels":["area:mcp"],"created":"2026-09-29T17:31:31Z","updated":"2026-10-16T13:12:33Z","closed":"2026-10-08T10:11:09Z","reactions":11,"thumbs_up":11,"comments":11,"severity":"high","severity_reasons":["title:breakage"],"topic":"install/update","classes":["install/update","mcp/tools","always-on/daemon","self-host/headless","crash/hang"],"cosmetic_ui":false,"summary":"Immediately after installing the latest macOS security update and rebooting, all Copilot CLI sessions—both newly started sessions and resumed existing sessions—became unable to process prompts. GitHub Copilot CLI 1.0.90-3. macOS Golden…","versions":["1.0.90-3.","27.0.1","1.0.90-3"],"error_snippets":["Failed to start MCP Servers: Error: The shared writer lock or its directory changed.","Scheduled queue processing failed; retrying once {\"error\":\"Message(\\\"The shared writer lock or its directory changed.\\\")\"}","Scheduled queue processing failed after retry; queue drain abandoned {\"error\":\"Message(\\\"The shared writer lock or its directory changed.\\\")\"}","Failed to start MCP Servers: Error: Shared writer storage I/O failed.","Request session.send failed with message: Shared writer storage I/O failed."],"workaround":null,"fixed_in":null,"score":82.6,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openclaw/openclaw#162448","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":162448,"title":"[Bug]: Linux desktop app minimize, maximize, and close buttons are overlapping side panel controls","url":"https://github.com/openclaw/openclaw/issues/162448","state":"open","state_reason":null,"labels":["bug","regression","P2","clawsweeper:no-new-fix-pr","clawsweeper:needs-maintainer-review","clawsweeper:needs-info","issue-rating: 🦐 gold shrimp","impact:ux-friction"],"created":"2026-10-01T05:52:48Z","updated":"2026-10-25T13:03:19Z","closed":null,"reactions":4,"thumbs_up":2,"comments":6,"severity":"high","severity_reasons":["label:regression"],"topic":"install/update","classes":["install/update","network"],"cosmetic_ui":false,"summary":"Regression (worked before, now fails) The buttons are overlapping and you can't use the sidebar buttons Start v2026.9.7 on Linux, I'm using Ubuntu 26.04, installed as a .deb Do something to open the side panel See the icons all…","versions":["2026.9.7","26.04"],"error_snippets":["The buttons are overlapping and you can't use the sidebar buttons","3. See the icons all overlapping, you can't use the sidebar icons"],"workaround":null,"fixed_in":null,"score":82.2,"first_seen":"2026-10-21T09:24:21Z"},{"id":"NousResearch/hermes-agent#143657","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":143657,"title":"[Bug]: macOS headless: gateway ensure aborts with service_manager_unavailable because launchctl print gui/<uid> returns 125","url":"https://github.com/NousResearch/hermes-agent/issues/143657","state":"open","state_reason":null,"labels":["type/bug","comp/cli","comp/gateway","P2","os/macos"],"created":"2026-10-23T16:54:31Z","updated":"2026-10-24T09:41:00Z","closed":null,"reactions":1,"thumbs_up":1,"comments":2,"severity":"medium","severity_reasons":["class:always-on/daemon","class:self-host/headless"],"topic":"always-on/daemon","classes":["always-on/daemon","self-host/headless","install/update"],"cosmetic_ui":false,"summary":"What happens: hermes -z \"…\" (and any entrypoint that ensures the gateway) fails with Error: Gateway inaccessible: servicemanagerunavailable when run without a GUI login session — e.g. from a system LaunchDaemon with UserName= , or via sudo…","versions":["0.22.3","0.22.5","0.22.2"],"error_snippets":["Error: Gateway inaccessible: service_manager_unavailable",", if the LaunchAgent plist does not exist, either skip the"],"workaround":null,"fixed_in":null,"score":82.1,"first_seen":"2026-10-25T09:11:37Z"},{"id":"google-gemini/gemini-cli#5397","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":5397,"title":"Github MCP OAuth","url":"https://github.com/google-gemini/gemini-cli/issues/5397","state":"closed","state_reason":"completed","labels":["priority/p2"],"created":"2025-08-02T04:16:08Z","updated":"2026-10-15T19:28:37Z","closed":"2025-09-19T18:21:21Z","reactions":12,"thumbs_up":11,"comments":15,"severity":"medium","severity_reasons":["engagement 27","class:auth/session","class:mcp/tools"],"topic":"auth/session","classes":["auth/session","mcp/tools","self-host/headless"],"cosmetic_ui":false,"summary":"Attempting to use the official HTTP MCP server from github: https://github.com/github/github-mcp-server I was able to get it to work with a Github Personal Access Token(PAT), with the following config: but /mcp auth githubCopilot results…","versions":["0.1.15-nightly.250801.6f7beb41"],"error_snippets":["Error connecting to MCP server 'githubCopilot': MCP server 'githubCopilot' requires authentication. Please configure OAuth or check server settings.","✕ Failed to authenticate with MCP server 'githubCopilot': Failed to discover OAuth configuration from MCP server","I expected the `/mcp auth githubCopilot` to autodiscover, open the browser, ask to authenticate, and hit the gemini-cli callback server to do the OAuth handshake."],"workaround":null,"fixed_in":null,"score":81.9,"first_seen":"2026-10-19T18:38:33Z"},{"id":"NousResearch/hermes-agent#118753","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":118753,"title":"Curator's automatic inactivity archival can silently break kanban swarm verifier/synthesizer (requesting-code-review, humanizer)","url":"https://github.com/NousResearch/hermes-agent/issues/118753","state":"open","state_reason":null,"labels":["type/bug","comp/agent","comp/cli","comp/cron","tool/skills","P2","needs-decision","area/profiles"],"created":"2026-09-22T03:46:14Z","updated":"2026-10-22T17:07:30Z","closed":null,"reactions":1,"thumbs_up":1,"comments":8,"severity":"medium","severity_reasons":["class:always-on/daemon","class:mcp/tools"],"topic":"always-on/daemon","classes":["always-on/daemon","mcp/tools","install/update","crash/hang"],"cosmetic_ui":false,"summary":"hermes kanban swarm's verifier role hardcodes the skill requesting-code-review (hermescli/kanbanswarm.py:238). This skill ships bundled (skills/software-development/requesting-code-review/), so it normally exists — but the Curator's…","versions":[],"error_snippets":["Error: Unknown skill(s): requesting-code-review","kanban.failure_limit",", and nothing prevents the exact same archival from recurring in another","Once archived, every subsequent `kanban swarm` verifier card crashes at launch with `Unknown skill(s): requesting-code-review`, hits `kanban.failure_limit` (default 2), and auto-blocks — with no board-visible explanation","`hermes curator pin` also cannot be used as a manual workaround for this specific skill: it explicitly refuses bundled/hub skills (*\"'requesting-code-review' is bundled or hub-installed — cannot pin (only agent-created s"],"workaround":null,"fixed_in":null,"score":81.6,"first_seen":"2026-10-18T04:32:26Z"},{"id":"anthropics/claude-code#102534","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":102534,"title":"[BUG] Desktop: Remote Control drops on every idle session and can't be re-enabled until a message is sent from the Mac","url":"https://github.com/anthropics/claude-code/issues/102534","state":"open","state_reason":null,"labels":["bug","has repro","platform:macos","area:desktop"],"created":"2026-10-15T19:02:07Z","updated":"2026-10-22T16:23:23Z","closed":null,"reactions":3,"thumbs_up":3,"comments":3,"severity":"medium","severity_reasons":["class:self-host/headless"],"topic":"self-host/headless","classes":["self-host/headless","network"],"cosmetic_ui":false,"summary":"In the Claude desktop app (Code tab), Remote Control only stays connected while a session's process is running. Once a session finishes its turn and goes idle, the app shuts its process down, and the Remote Control link drops with it. The…","versions":["2.1.300"],"error_snippets":["In the Claude desktop app (Code tab), Remote Control only stays connected while a session's process is running. Once a session finishes its turn and goes idle, the app shuts its process down, and the Remote Control link ","Trying to turn Remote Control back on for an idle session through the app's session tool (`set_remote_control`) is refused:","5. Back on the desktop, the session's Remote Control state reads \"off\". Turning it on is refused until a message is sent from the desktop.","Related but different: #33041 (CLI Remote Control disconnects) and #95339 (forked desktop sessions don't auto-connect). This one is about every idle desktop session losing Remote Control."],"workaround":null,"fixed_in":null,"score":81.3,"first_seen":"2026-10-22T09:27:06Z"},{"id":"anthropics/claude-code#102116","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":102116,"title":"Desktop app: archived sessions can't be shown in the sidebar anymore","url":"https://github.com/anthropics/claude-code/issues/102116","state":"closed","state_reason":"not_planned","labels":["bug","platform:windows","regression","area:desktop"],"created":"2026-10-14T17:16:51Z","updated":"2026-10-26T08:10:41Z","closed":"2026-10-21T12:36:52Z","reactions":13,"thumbs_up":13,"comments":12,"severity":"high","severity_reasons":["label:regression"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"The filter button next to \"+\" at the top of the session list used to let you show archived sessions. After the update it doesn't. The only way to them now is \"View All\", which opens a whole separate panel listing every session. You archive…","versions":[],"error_snippets":["The filter button next to \"+\" at the top of the session list used to let you show archived sessions. After the update it doesn't. The only way to them now is \"View All\", which opens a whole separate panel listing every s"],"workaround":null,"fixed_in":null,"score":81.0,"first_seen":"2026-10-21T09:24:21Z"},{"id":"NousResearch/hermes-agent#122222","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":122222,"title":"[Bug]: cron external worker cannot import dependencies on self-managed installs — every scheduled job fails before its ownership ack","url":"https://github.com/NousResearch/hermes-agent/issues/122222","state":"closed","state_reason":"completed","labels":["type/bug","comp/cron","P1","sweeper:risk-compatibility","area/install-update"],"created":"2026-09-25T03:38:39Z","updated":"2026-10-13T15:07:08Z","closed":"2026-09-28T22:02:39Z","reactions":4,"thumbs_up":3,"comments":34,"severity":"high","severity_reasons":["label:p1"],"topic":"always-on/daemon","classes":["always-on/daemon","install/update"],"cosmetic_ui":false,"summary":"On a self-managed (shell-installer / PM) Hermes install, every scheduled cron job fails before it can run. The restart-safe external worker is spawned as sys.executable -m cron.scheduler with a sanitized environment whose PYTHONPATH…","versions":[],"error_snippets":["Restart-safe cron worker dispatch failed: cron external worker exited before","ownership acknowledgement (exit 1); worker stderr:","ModuleNotFoundError: No module named 'ruamel'","STORE=\"$(hermes gateway status --json 2>/dev/null | python3 -c 'import json,sys;print(json.load(sys.stdin)[\"python\"])' 2>/dev/null)\" \\","|| STORE=\"$HERMES_HOME/tools/python-\"*/bin/python3 # any launcher Python; the point is it exposes no deps"],"workaround":null,"fixed_in":null,"score":80.3,"first_seen":"2026-10-18T04:29:32Z"},{"id":"kirodotdev/Kiro#5873","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":5873,"title":"MCP tools not injected into LLM tool list for workspace-defined agents (.kiro/agents/*.json)","url":"https://github.com/kirodotdev/Kiro/issues/5873","state":"open","state_reason":null,"labels":["pending-maintainer-response","cli"],"created":"2026-02-19T21:00:28Z","updated":"2026-09-16T10:54:09Z","closed":null,"reactions":7,"thumbs_up":3,"comments":9,"severity":"medium","severity_reasons":["engagement 16","class:mcp/tools"],"topic":"mcp/tools","classes":["mcp/tools","self-host/headless"],"cosmetic_ui":false,"summary":"Any agent defined via .kiro/agents/.json does not receive MCP tools in its tool list. Only the built-in kirodefault agent receives MCP tools. MCP servers initialize successfully in all cases, but tools are only injected for the built-in…","versions":["1.26.2"],"error_snippets":["All tests run from the same workspace directory. MCP servers initialized successfully (✓) in every case.","Workspace-level agent customization (`prompt`, `resources`, `allowedTools`, `hooks`) cannot be used with MCP tools. This blocks any workflow that requires both custom agent behavior and MCP server access."],"workaround":{"text":"`kiro-cli chat --agent kiro_default` The built-in `kiro_default` agent correctly receives MCP tools from global `~/.kiro/settings/mcp.json`. Custom agent prompt/resources must be loaded manually at session start.","source":"issue body","url":"https://github.com/kirodotdev/Kiro/issues/5873"},"fixed_in":null,"score":80.1,"first_seen":"2026-10-19T18:38:33Z"},{"id":"NousResearch/hermes-agent#122183","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":122183,"title":"[Bug]: Windows gateway on the PM runtime prepends the pre-PM venv and crashes hosted_room_worker (No module named 'pydantic_core._pydantic_core')","url":"https://github.com/NousResearch/hermes-agent/issues/122183","state":"closed","state_reason":"completed","labels":["type/bug","comp/gateway","P1","sweeper:risk-message-delivery","sweeper:risk-compatibility","sweeper:risk-platform-windows","area/install-update","os/windows"],"created":"2026-09-25T03:09:14Z","updated":"2026-10-17T10:21:03Z","closed":"2026-09-26T14:27:58Z","reactions":2,"thumbs_up":2,"comments":24,"severity":"high","severity_reasons":["label:p1","title:breakage"],"topic":"always-on/daemon","classes":["always-on/daemon","install/update","crash/hang"],"cosmetic_ui":false,"summary":"A Windows install that moved to the PM runtime (managed Python 3.14, dependencies under installs/ /environments/ /venv) still has the old pre-PM \\venv (Python 3.11) on disk. After hermes update --gateway respawned the messaging gateway…","versions":["10.0.26200","3.14.7"],"error_snippets":["ERROR gateway.run: Supervised task hosted_room_worker died: ModuleNotFoundError(\"No module named 'pydantic_core._pydantic_core'\")","ERROR gateway.run: Supervised task hosted_room_worker died 5 times in rapid succession (each within 300s of restart) — giving up restarts","File \"<install>\\gateway\\run_startup.py\", line 794, in _start_hosted_room_worker_sync","ModuleNotFoundError: No module named 'pydantic_core._pydantic_core'","PYTHONPATH=<install>;<install>\\venv\\Lib\\site-packages;<install>;<HERMES_HOME>\\installs\\<id>\\environments\\<gen>\\venv\\Lib\\site-packages"],"workaround":null,"fixed_in":null,"score":79.7,"first_seen":"2026-10-18T04:32:26Z"},{"id":"anthropics/claude-code#96105","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":96105,"title":"[BUG] macOS 12: Desktop app stuck on bundled Claude Code 2.1.260, Opus 5.5 greyed out — yet 2.1.280 lists this app's SDK wrapper (0.3.260) as tested while `stable` (2.1.267) does not","url":"https://github.com/anthropics/claude-code/issues/96105","state":"open","state_reason":null,"labels":["bug","platform:macos","area:packaging","area:desktop"],"created":"2026-09-22T18:03:50Z","updated":"2026-09-25T14:23:32Z","closed":null,"reactions":7,"thumbs_up":7,"comments":6,"severity":"medium","severity_reasons":["class:crash/hang"],"topic":"crash/hang","classes":["crash/hang","install/update"],"cosmetic_ui":false,"summary":"On macOS 12 (Monterey, Intel), the Desktop app's model picker shows Opus 5.5 greyed out with the tooltip \"Update Claude to use this model on this computer\", while the app's own update check reports \"You're using the latest version.\" The…","versions":["2.1.260","2.1.280","0.3.260","2.1.267"],"error_snippets":["The machine is stranded between two update paths that cannot meet:","That is the crux: **2.1.280 explicitly lists this app's wrapper version as tested** (same `harnessSchema`), while the `stable` build the app follows does not. The app therefore never advances past the 2.1.260 it was ship","macOS 12 users are silently cut off from new models (Opus 5.5 today, whatever follows next) in the Desktop app, with no path forward and no actionable message: the picker says \"update Claude\", and updating Claude answers","3. Failing either, make the tooltip honest for OS-stranded installs (\"this model needs Desktop x.y, which requires macOS 13\") instead of pointing at an update that does not exist for that machine."],"workaround":null,"fixed_in":null,"score":79.4,"first_seen":"2026-10-18T04:33:44Z"},{"id":"google-gemini/gemini-cli#8132","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":8132,"title":"The input token count (xxx) exceeds the maximum number of tokens allowed (1048576) in service logs","url":"https://github.com/google-gemini/gemini-cli/issues/8132","state":"closed","state_reason":"duplicate","labels":["priority/p2","area/agent","status/bot-triaged","kind/bug","status/need-information"],"created":"2025-09-09T22:34:33Z","updated":"2026-09-13T07:10:23Z","closed":"2025-10-02T21:58:29Z","reactions":6,"thumbs_up":6,"comments":24,"severity":"medium","severity_reasons":["engagement 30","class:always-on/daemon","class:auth/session"],"topic":"always-on/daemon","classes":["always-on/daemon","auth/session","network"],"cosmetic_ui":false,"summary":"(Note: this is copy/pasted from google-internal bug https://buganizer.corp.google.com/issues/429600201#comment13, with internal links removed... they're still available on buganzier, though.) ------------- We are seeing a large number of…","versions":[],"error_snippets":["The input token count (4417490) exceeds the maximum number of tokens allowed (1048576).","We are seeing a large number of errors in the CCPA / PredictionService logs where","2. Call to PredictionService.StreamGenerateChat, which calls through to PredictionService.StreamGenerateContent. (Which sometimes fails with something like `The input token count (4417490) exceeds the maximum number of t","3. Gemini CLI is not properly handling an error response from CountTokens. (e.g. if the CountTokens call fails, does Gemini CLI abort, or does it just proceed without trying to do any truncation?)","Can you please confirm the Gemini CLI behavior around handling any CountTokens errors?"],"workaround":null,"fixed_in":null,"score":79.4,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openclaw/openclaw#88312","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":88312,"title":"[Bug]: [Regression] 2026.5.27: Codex app-server turn-completion stall returns (\"Codex stopped before confirming the turn was complete\") — regression of #84076 (fixed by #85107)","url":"https://github.com/openclaw/openclaw/issues/88312","state":"closed","state_reason":"completed","labels":["bug","regression","P1","clawsweeper:no-new-fix-pr","clawsweeper:fix-shape-clear","clawsweeper:needs-maintainer-review","clawsweeper:needs-product-decision","clawsweeper:needs-live-repro","impact:session-state","impact:message-loss","issue-rating: 🐚 platinum hermit"],"created":"2026-05-30T11:28:42Z","updated":"2026-09-14T03:48:36Z","closed":"2026-07-18T01:23:58Z","reactions":5,"thumbs_up":5,"comments":22,"severity":"high","severity_reasons":["label:p1","title:breakage"],"topic":"self-host/headless","classes":["self-host/headless","install/update","always-on/daemon","mcp/tools","network"],"cosmetic_ui":false,"summary":"Regression (worked before, now fails) On the Codex app-server (ChatGPT Plus sub), a multi-tool agent turn reliably fails with Codex stopped before confirming the turn was complete starting in 2026.5.27; 2026.5.26 completes the same turns…","versions":["2026.5.27","2026.5.26","2026.5.28-beta.","5.28-beta.1"],"error_snippets":["Bisect: 2026.5.26 ✅ (stopReason: stop, ~23 tool calls) → 2026.5.27 ❌ → 5.28-beta.1 ❌ → 5.28-beta.4 ❌. From production cron history: same job completed on 5.26, failed the first run after the 5.27 upgrade and every run si","Codex stops without saying the turn is complete or emitting an error."],"workaround":{"text":"Only found workaround is pinning 5.26.","source":"issue body","url":"https://github.com/openclaw/openclaw/issues/88312"},"fixed_in":null,"score":79.0,"first_seen":"2026-10-18T04:30:03Z"},{"id":"github/copilot-cli#4870","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":4870,"title":"MCP: Figma remote server (mcp.figma.com) fails to load — `-32601` on `server/discover` treated as fatal (works in VS Code)","url":"https://github.com/github/copilot-cli/issues/4870","state":"closed","state_reason":"completed","labels":[],"created":"2026-09-16T08:51:26Z","updated":"2026-09-29T12:57:06Z","closed":"2026-09-20T18:33:29Z","reactions":12,"thumbs_up":12,"comments":8,"severity":"medium","severity_reasons":["engagement 20","class:self-host/headless","class:mcp/tools"],"topic":"self-host/headless","classes":["self-host/headless","mcp/tools","always-on/daemon","install/update","crash/hang","auth/session"],"cosmetic_ui":false,"summary":"The Figma hosted MCP server (https://mcp.figma.com/mcp) authenticates and initializes successfully, but the CLI never registers its tools. The CLI's discovery probe receives -32601 Method not found, records it as a fatal server failure…","versions":["1.0.85"],"error_snippets":["[DEBUG] rt_mcp::native_host: Recorded failure for server figma: MCP error -32601: Method not found.","[WARNING] copilot_runtime::session::mcp::agent_host: MCP error -32601: Method not found. { server: \"figma\" }","[ERROR] rmcp::transport::worker: worker quit with fatal: Transport channel closed,","when Client(OAuthChallenge { ... mcp.figma.com ... status_code: 401 })","-32601 Method not found"],"workaround":{"text":"Update to a newer CLI build that supports Figma's `2026-07-28` MCP protocol handshake.","source":"issue body","url":"https://github.com/github/copilot-cli/issues/4870"},"fixed_in":null,"score":79.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openclaw/openclaw#161378","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":161378,"title":"Severe Gateway/SQLite regression across 2026.8→2026.9: after two months of freezes and lock failures we rolled back to 2026.7.35 LTS to get a working assistant","url":"https://github.com/openclaw/openclaw/issues/161378","state":"open","state_reason":null,"labels":["clawsweeper:no-new-fix-pr","clawsweeper:needs-maintainer-review","impact:crash-loop","P0","issue-rating: 🦪 silver shellfish","impact:ux-release-blocker"],"created":"2026-09-29T20:42:40Z","updated":"2026-10-19T18:27:17Z","closed":null,"reactions":2,"thumbs_up":1,"comments":5,"severity":"critical","severity_reasons":["label:p0"],"topic":"always-on/daemon","classes":["always-on/daemon","install/update","crash/hang"],"cosmetic_ui":false,"summary":"We use OpenClaw daily — multiple agents, scheduled cron jobs, and a long-lived state directory on Windows. We just spent two months fighting it, and only after rolling back from 2026.9.6 to 2026.7.35 (Extended-Stable LTS) did we get a…","versions":["2026.7.35","2026.9.6","9.6","2.0"],"error_snippets":["busy_timeout=5000","SQLite transaction admission was refused","after rolling back from 2026.9.6 to 2026.7.35 (Extended-Stable LTS) did we get","\"boiling frog\" — each upgrade felt only slightly worse than the last, so we","to daily users. Shipping this level of lock contention, startup failure, and"],"workaround":null,"fixed_in":null,"score":78.5,"first_seen":"2026-10-18T04:33:05Z"},{"id":"omacom/omarchy#11727","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":11727,"title":"Shell (quickshell) SIGSEGV when a click reassigns a Repeater's array model — crash in VDMListDelegateDataType::createMissingProperties","url":"https://github.com/omacom/omarchy/issues/11727","state":"open","state_reason":null,"labels":["bug"],"created":"2026-09-13T18:18:01Z","updated":"2026-10-24T21:29:20Z","closed":null,"reactions":1,"thumbs_up":1,"comments":12,"severity":"high","severity_reasons":["title:breakage"],"topic":"crash/hang","classes":["crash/hang","install/update"],"cosmetic_ui":false,"summary":"The Omarchy shell (quickshell) segfaulted at 19:57:33 while I clicked on its UI (bar/panel element). Quickshell's in-process crash recovery restarted the engine (\"Quickshell has been restarted\") and the bar came back by itself. No data…","versions":["6.11.2","0.3.1-1","4.0.3-1","6.11.2-3"],"error_snippets":["→ QQmlBoundSignalExpression::evaluate (×3 nested signal/binding layers)","→ QV4::Module::evaluate → QV4::Runtime::TypeofName::call","(bar/panel element). Quickshell's in-process crash recovery restarted the engine","The crash happened while the engine was handling a mouse release on a shell","Clicking a bar/panel widget should never take down the shell engine."],"workaround":null,"fixed_in":null,"score":78.4,"first_seen":"2026-10-25T09:11:37Z"},{"id":"omacom/omarchy#5755","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":5755,"title":"Screen on laptop not turning back on after suspend since 3.8 upgrade","url":"https://github.com/omacom/omarchy/issues/5755","state":"open","state_reason":null,"labels":["bug","compatibility"],"created":"2026-05-11T11:20:14Z","updated":"2026-10-08T01:59:37Z","closed":null,"reactions":6,"thumbs_up":3,"comments":7,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update","always-on/daemon","crash/hang","auth/session"],"cosmetic_ui":false,"summary":"Dell Pro Max 16 Worked well up until 3.8 upgrade today, now the screen wont turn back on after suspend (have to hard reset): Anyone else seeing this?/","versions":["3.8","0.54.3","1.4.2","24.1.11"],"error_snippets":["Info: 16-core Intel Core Ultra 7 265H [MCP] speed (MHz): avg: 400","min/max: 400/5300:4600:2500","API: EGL Message: EGL data requires eglinfo. Check --recommends.","May 11 12:54:34 Laptop fprintd[8491]: Device reported an error during verify: Cannot run while suspended.","Anyone else seeing this?/"],"workaround":null,"fixed_in":null,"score":78.3,"first_seen":"2026-10-19T18:38:33Z"},{"id":"omacom/omarchy#2415","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":2415,"title":"Intel based Macbook keyboard doesn't respond on installation","url":"https://github.com/omacom/omarchy/issues/2415","state":"open","state_reason":null,"labels":["bug","mac","compatibility"],"created":"2025-10-12T22:59:44Z","updated":"2026-10-07T22:25:16Z","closed":null,"reactions":7,"thumbs_up":7,"comments":8,"severity":"medium","severity_reasons":["engagement 15","class:install/update"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"MacBook Pro 13\" (2020) with an Intel chip (1,4 GHz Quad-Core Intel Core i5) I downloaded the Omarchy (v3.0.2) ISO from the website and created a bootable USB using this guide. It boots successfully — I see the Omarchy logo when the…","versions":["3.0.2"],"error_snippets":["I downloaded the Omarchy (v3.0.2) ISO from the website and created a bootable USB using this guide. It boots successfully — I see the Omarchy logo when the installation starts — but at the first step (language selection)"],"workaround":null,"fixed_in":null,"score":77.6,"first_seen":"2026-10-19T18:38:33Z"},{"id":"omacom/omarchy#4801","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":4801,"title":"Speaker output not working in Asus Zephyrus G14 after 3.4 update","url":"https://github.com/omacom/omarchy/issues/4801","state":"open","state_reason":null,"labels":["bug"],"created":"2026-02-28T10:41:07Z","updated":"2026-10-16T07:41:56Z","closed":null,"reactions":7,"thumbs_up":7,"comments":8,"severity":"medium","severity_reasons":["engagement 15","class:install/update"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"When updated to lastest veriosn of Omarchy (3.4). The native laptop speaker is not working, which was working fine pre-update. Bluetooth works fine, and other external audio peripheral works good","versions":["3.4"],"error_snippets":["When updated to lastest veriosn of Omarchy (3.4). The native laptop speaker is not working, which was working fine pre-update. Bluetooth works fine, and other external audio peripheral works good"],"workaround":null,"fixed_in":null,"score":77.6,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openclaw/openclaw#108435","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":108435,"title":"[Bug]: update to openclaw 2026.7.1: gateway fails to start w/ error","url":"https://github.com/openclaw/openclaw/issues/108435","state":"closed","state_reason":"completed","labels":["bug","regression","clawsweeper:needs-info","impact:crash-loop","P0","issue-rating: 🦪 silver shellfish","maturity:stable","impact:ux-release-blocker"],"created":"2026-07-15T19:13:32Z","updated":"2026-09-14T01:18:32Z","closed":"2026-09-14T01:18:32Z","reactions":7,"thumbs_up":3,"comments":15,"severity":"critical","severity_reasons":["label:p0"],"topic":"always-on/daemon","classes":["always-on/daemon","install/update","crash/hang"],"cosmetic_ui":false,"summary":"openclaw-2026-07-15.log Regression (worked before, now fails) gateway doesn't start with manual launch Error: gateway did not start on 127.0.0.1:18789 Pls have a look! update to openclaw 2026.7.1 openclaw gateway run --force CLI is…","versions":["2026.7.1"],"error_snippets":["Error: gateway did not start on 127.0.0.1:18789","`Error: gateway did not start on 127.0.0.1:18789`"],"workaround":null,"fixed_in":null,"score":77.3,"first_seen":"2026-10-18T04:30:03Z"},{"id":"NousResearch/hermes-agent#122349","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":122349,"title":"Plugin runtime state inside a member folder causes an infinite sync/rebuild/re-exec loop","url":"https://github.com/NousResearch/hermes-agent/issues/122349","state":"open","state_reason":null,"labels":["type/bug","comp/cli","comp/plugins","P3","sweeper:risk-compatibility","area/install-update"],"created":"2026-09-25T05:52:16Z","updated":"2026-10-06T02:32:24Z","closed":null,"reactions":1,"thumbs_up":1,"comments":9,"severity":"medium","severity_reasons":["class:install/update","class:mcp/tools"],"topic":"install/update","classes":["install/update","mcp/tools","always-on/daemon","crash/hang"],"cosmetic_ui":false,"summary":"pm/workspace.py::membersstamp() hashes every file in each enabled plugin member directory to decide whether the recorded dependency environment is current (pm/install.py::venviscurrent). The only exclusions are the hard-coded MEMBEREXCLUDE…","versions":[],"error_snippets":[") and names ending in","~/.hermes/installs/<id>/environments/","`pm/workspace.py::members_stamp()` hashes every file in each enabled plugin","member directory to decide whether the recorded dependency environment is","current (`pm/install.py::venv_is_current`). The only exclusions are the"],"workaround":null,"fixed_in":null,"score":77.2,"first_seen":"2026-10-18T04:32:26Z"},{"id":"github/copilot-cli#5192","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":5192,"title":"ACP mode: MCP OAuth token is never persisted after successful sign-in — every new session re-opens the browser","url":"https://github.com/github/copilot-cli/issues/5192","state":"open","state_reason":null,"labels":["area:non-interactive","area:mcp"],"created":"2026-10-19T15:49:22Z","updated":"2026-10-21T13:01:30Z","closed":null,"reactions":1,"thumbs_up":1,"comments":2,"severity":"medium","severity_reasons":["class:auth/session","class:mcp/tools"],"topic":"auth/session","classes":["auth/session","mcp/tools","always-on/daemon","self-host/headless","install/update"],"cosmetic_ui":false,"summary":"In ACP mode (copilot --acp), the OAuth token for a remote HTTP MCP server is never written to disk or the OS keychain after a successful sign-in. Each new ACP session (for example, each new thread in an ACP client) starts unauthenticated…","versions":["1.0.98"],"error_snippets":["14:32:47.302Z [WARNING] [rust:copilot_runtime::session::mcp::agent_host] HTTP 401 challenge (...) {\"server\":\"atlassian-cloud\"}","15:22:30Z [WARNING] [rust:copilot_runtime::session::mcp::agent_host] HTTP 401 challenge (...) {\"server\":\"atlassian-cloud\"}","15:23:48Z ... timed out after 60s","In ACP mode (`copilot --acp`), the OAuth token for a remote HTTP MCP server is never written to disk or the OS keychain after a successful sign-in. Each new ACP session (for example, each new thread in an ACP client) sta","Evidence from a successful sign-in (macOS, Atlassian remote MCP `https://mcp.atlassian.com/v2/mcp`):"],"workaround":null,"fixed_in":null,"score":76.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"continuedev/continue#3872","harness":"continue","harness_name":"Continue","repo":"continuedev/continue","number":3872,"title":"Unable to use Jetbrains gateway.","url":"https://github.com/continuedev/continue/issues/3872","state":"closed","state_reason":"not_planned","labels":["area:indexing","kind:bug","ide:jetbrains","priority:medium","stale","os:windows","os:linux"],"created":"2025-01-28T06:38:06Z","updated":"2026-10-03T01:49:39Z","closed":"2026-10-03T01:49:39Z","reactions":12,"thumbs_up":12,"comments":21,"severity":"medium","severity_reasons":["engagement 33","class:always-on/daemon"],"topic":"always-on/daemon","classes":["always-on/daemon","self-host/headless","install/update","crash/hang","auth/session","mcp/tools"],"cosmetic_ui":false,"summary":"I believe this is a bug. I'll try to join the Continue Discord for questions I'm not able to find an open issue that reports the same bug I've seen the troubleshooting guide on the Continue Docs When I open a project through Jetbrains…","versions":["0.0.86","2024.3.1"],"error_snippets":["[2025-01-28T03:31:28] (node:64460) ExperimentalWarning: The Fetch API is an experimental feature. This feature could change at any time","[2025-01-28T03:31:35] error when indexing: Error: Failed to generate embeddings for 401 chunks with provider: A::all-MiniLM-L6-v2::500: Error: `local_files_only=true` or `env.allowRemoteModels=false` and file was not fou","[2025-01-28T06:31:37] (node:69712) ExperimentalWarning: The Fetch API is an experimental feature. This feature could change at any time","When I open a project through Jetbrains gateway, I can't get the file list in Continue."],"workaround":null,"fixed_in":null,"score":76.4,"first_seen":"2026-10-19T18:38:33Z"},{"id":"NousResearch/hermes-agent#122402","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":122402,"title":"[Bug]: Ubuntu historical takeover fails building python-olm when managed Python requires missing clang++","url":"https://github.com/NousResearch/hermes-agent/issues/122402","state":"open","state_reason":null,"labels":["type/bug","comp/cli","platform/matrix","P2","sweeper:risk-compatibility","area/install-update"],"created":"2026-09-25T06:52:13Z","updated":"2026-10-23T15:11:57Z","closed":null,"reactions":1,"thumbs_up":1,"comments":14,"severity":"medium","severity_reasons":["engagement 15","class:install/update"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"A git-installed Hermes update on Ubuntu 24.04 failed while preparing the new isolated dependency environment during the historical PM takeover. The selected features included matrix; mautrix[encryption] pulled python-olm==3.2.16, which…","versions":["3.2.16","0.21.4","0.20.5","0.0.0"],"error_snippets":["✗ Installing Python dependencies failed","building '_libolm' extension","error: [Errno 2] No such file or directory: 'clang++'","hint: `python-olm` (v3.2.16) was included because `hermes-agent[matrix]` (v0.0.0) depends on `mautrix[encryption]` (v0.21.1) which depends on `python-olm`","Update preparation failed: venv: uv sync exited 1"],"workaround":null,"fixed_in":null,"score":76.1,"first_seen":"2026-10-18T04:32:26Z"},{"id":"anthropics/claude-code#95975","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":95975,"title":"[BUG] Claude in Chrome 1.0.94: side panel shows \"Can't reach the Claude extension.\" in Vivaldi; classic side panel opt-out removed","url":"https://github.com/anthropics/claude-code/issues/95975","state":"open","state_reason":null,"labels":["bug","has repro","platform:macos","regression","area:browser-extension","area:chrome"],"created":"2026-09-22T05:43:58Z","updated":"2026-10-25T08:48:35Z","closed":null,"reactions":5,"thumbs_up":5,"comments":5,"severity":"high","severity_reasons":["label:regression"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"This is about the Claude in Chrome browser extension, not the Claude Code CLI. In Vivaldi, the extension's new side panel (the embedded claude.ai Cowork interface) never works. It only shows: Up to extension 1.0.93, you could work around…","versions":["1.0.93","1.0.94"],"error_snippets":["Side panel text: Can't reach the Claude extension.","In Vivaldi, the extension's new side panel (the embedded claude.ai Cowork interface) never works. It only shows:","> Can't reach the Claude extension.","Up to extension **1.0.93**, you could work around this in the extension options: **Side panel experience → New side panel experience → off**. That brought back the classic side panel, which works fine in Vivaldi.","According to the 1.0.94 code, the classic panel is still used when `sidepanel.html` is opened with `?mode=window`, so the classic UI hasn't been removed. It just can't be chosen for the side panel anymore."],"workaround":null,"fixed_in":null,"score":76.0,"first_seen":"2026-10-24T08:54:58Z"},{"id":"aaif-goose/goose#12068","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":12068,"title":"Support SEP-2640 Skills Extension (skills over MCP) as host","url":"https://github.com/aaif-goose/goose/issues/12068","state":"open","state_reason":null,"labels":[],"created":"2026-09-14T21:17:52Z","updated":"2026-10-19T18:55:00Z","closed":null,"reactions":1,"thumbs_up":1,"comments":13,"severity":"medium","severity_reasons":["class:mcp/tools"],"topic":"mcp/tools","classes":["mcp/tools","self-host/headless","install/update","crash/hang"],"cosmetic_ui":false,"summary":"What problem would this solve? MCP has no standard way to ship workflow instructions alongside the tools they describe. Server instructions are size-bounded, so complex workflows do not fit. Today a server and the skill that teaches how to…","versions":[],"error_snippets":["MCP has no standard way to ship workflow instructions alongside the tools they describe. Server instructions are size-bounded, so complex workflows do not fit. Today a server and the skill that teaches how to use it are ","SEP-2640 (io.modelcontextprotocol/skills, modelcontextprotocol/modelcontextprotocol#2640, merged) fixes this by serving Agent Skills over MCP Resources: skill:// URIs plus skills/list, skills/get, and optional resources/","Goose as host supports the Skills Extension:"],"workaround":null,"fixed_in":null,"score":75.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"github/copilot-cli#5253","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":5253,"title":"CLI 1.0.103: persistent session-lock timeouts and unresponsive tasks amid host event acknowledgement failures","url":"https://github.com/github/copilot-cli/issues/5253","state":"open","state_reason":null,"labels":["area:sessions","area:agents"],"created":"2026-10-25T15:55:58Z","updated":"2026-10-26T05:29:08Z","closed":null,"reactions":0,"thumbs_up":0,"comments":2,"severity":"medium","severity_reasons":["class:crash/hang","class:network"],"topic":"crash/hang","classes":["crash/hang","network","install/update"],"cosmetic_ui":false,"summary":"A long-lived Copilot CLI session becomes largely unresponsive. Opening/refreshing tasks shows: The process stays alive, and some nested tool activity continues, but the UI/event reader repeatedly fails to acquire the session lock. macOS…","versions":["1.0.103","1.0.96","25.472","51.503"],"error_snippets":["Failed to refresh tasks: Error: session lock unavailable during is_session_generation_current: acquisition timed out","process-1792833427506-51062.1.log:L1 2026-10-25T13:48:30.453Z [WARNING] [rust:copilot_runtime::session::pending_request_flow] pending-request event was not delivered to the session host {\"error\": \"GenericFailure, session","process-1792833427506-51062.1.log:L39 2026-10-25T13:48:30.777Z [ERROR] SessionClient poll loop error (will retry): Error: session lock unavailable during is_session_generation_current: acquisition timed out","process-1792833427506-51062.1.log:L41 2026-10-25T13:48:31.076Z [WARNING] [rust:otel] dropping OTel tool facts: pending call limit exceeded","process-1792833427506-51062.1.log:L47 2026-10-25T13:48:31.091Z [WARNING] [rust:events_engine::host_delivery] event was not delivered to its registered session host {\"error\": \"GenericFailure, session event delivery failed"],"workaround":null,"fixed_in":null,"score":75.2,"first_seen":"2026-10-26T10:12:02Z"},{"id":"github/copilot-cli#4935","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":4935,"title":"Built-in Slack MCP integration requests full scope superset even when only read tools are exposed","url":"https://github.com/github/copilot-cli/issues/4935","state":"open","state_reason":null,"labels":["area:mcp"],"created":"2026-09-22T12:51:56Z","updated":"2026-10-26T05:28:48Z","closed":null,"reactions":5,"thumbs_up":4,"comments":1,"severity":"medium","severity_reasons":["class:mcp/tools"],"topic":"mcp/tools","classes":["mcp/tools","self-host/headless","auth/session"],"cosmetic_ui":false,"summary":"When authenticating the built-in Slack MCP server ( mcp.slack.com ) via Copilot CLI, the generated OAuth consent URL requests the full superset of Slack scopes — including write scopes ( chat:write , files:write , channels:write …","versions":["1.0.87-0"],"error_snippets":["When authenticating the built-in Slack MCP server ( mcp.slack.com ) via Copilot CLI, the generated OAuth consent URL requests the full superset of Slack scopes — including write scopes ( chat:write , files:write , channe","Example generated authorize URL (redacted):","Expected behavior","Users are asked to grant far broader permissions (including message sending, file uploads, channel creation, list/canvas writes) than the tool set they can actually use, violating least-privilege expectations.","1. Ensure the built-in Slack MCP server ( mcp.slack.com ) is available in Copilot CLI (no custom config needed — it's a first-party integration)."],"workaround":null,"fixed_in":null,"score":75.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"anthropics/claude-code#102135","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":102135,"title":"[BUG] Desktop 2.37337.1 (Linux): \"Local environment is no longer available\" toast on every new session from a project group","url":"https://github.com/anthropics/claude-code/issues/102135","state":"closed","state_reason":"completed","labels":["bug","platform:linux","regression","area:desktop"],"created":"2026-10-14T18:27:01Z","updated":"2026-10-15T17:54:36Z","closed":"2026-10-15T12:00:39Z","reactions":17,"thumbs_up":17,"comments":6,"severity":"high","severity_reasons":["label:regression"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"After upgrading claude-desktop from 2.31226.1 to 2.37337.1 (apt, stable) on Ubuntu 26.04 (GNOME/Wayland), every click on the \"+\" next to a project group in the Code tab sidebar shows a toast: \"Couldn't switch environment: the Local…","versions":[],"error_snippets":["\"worker\":{\"type\":\"environment\",\"id\":\"__local__\"}","After upgrading claude-desktop from 2.31226.1 to 2.37337.1 (apt, stable) on Ubuntu 26.04 (GNOME/Wayland), every click on the \"+\" next to a project group in the Code tab sidebar shows a toast: \"Couldn't switch environment"],"workaround":null,"fixed_in":null,"score":74.1,"first_seen":"2026-10-18T04:33:44Z"},{"id":"kirodotdev/Kiro#11777","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":11777,"title":"kiro-cli headless with API key: GetProfile returns \"User is not authorized to access this feature\", governance fails, MCP and web tools disabled (Pro, Google sign-in)","url":"https://github.com/kirodotdev/Kiro/issues/11777","state":"open","state_reason":null,"labels":["auth","pending-maintainer-response","cli"],"created":"2026-10-04T02:32:53Z","updated":"2026-10-23T17:04:56Z","closed":null,"reactions":0,"thumbs_up":0,"comments":5,"severity":"medium","severity_reasons":["class:self-host/headless","class:auth/session"],"topic":"self-host/headless","classes":["self-host/headless","auth/session","mcp/tools"],"cosmetic_ui":false,"summary":"Linux (Ubuntu) cloud server in Tokyo kiro-cli 2.27.1 Account: Kiro Pro, individual account, signed up with Google (Gmail) sign-in. No organization or administrator. kiro-cli 2.27.1, Linux (Ubuntu) cloud server in Tokyo Headless mode, auth…","versions":["2.27.1"],"error_snippets":["kiro-cli chat --agent-engine v2 --no-interactive --output-format stream-json --model claude-opus-5.5","Chat works and the model answers. But at every start the CLI calls GetProfile for the API key and it fails on all endpoints:","1. About 20 seconds of startup delay on every `kiro-cli chat` invocation, spent retrying GetProfile across all endpoints (including unreachable ISO/GovCloud endpoints).","4. Search /tmp/chat.log for \"GetProfile failed\" and \"governance\".","5. Observe: AccessDeniedError \"User is not authorized to access this feature.\" on us-east-1, \"Invalid token\" on other regions, i/o errors on ISO endpoints, then \"MCP disabled, web tools disabled\"."],"workaround":null,"fixed_in":null,"score":74.1,"first_seen":"2026-10-19T18:38:33Z"},{"id":"NousResearch/hermes-agent#142336","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":142336,"title":"[Bug]: Desktop: reasoning-effort pill spins forever when opening a messaging-platform session","url":"https://github.com/NousResearch/hermes-agent/issues/142336","state":"open","state_reason":null,"labels":["type/bug","comp/gateway","platform/wecom","P2","area/sessions"],"created":"2026-10-21T02:43:15Z","updated":"2026-10-25T03:31:41Z","closed":null,"reactions":1,"thumbs_up":1,"comments":3,"severity":"medium","severity_reasons":["class:always-on/daemon"],"topic":"always-on/daemon","classes":["always-on/daemon"],"cosmetic_ui":false,"summary":"Opening a conversation that belongs to a messaging platform (Telegram DM/group, WeChat, Feishu, …) in Hermes Desktop leaves the composer's reasoning-effort pill in its loading state indefinitely. It never paints the chat's actual level…","versions":["0.22.2+2953.g0b8dee0"],"error_snippets":["but no effort column, and the desktop tile path reads","Opening a conversation that belongs to a messaging platform (Telegram DM/group, WeChat, Feishu, …) in Hermes Desktop leaves the composer's reasoning-effort pill in its loading state indefinitely. It never paints the chat","3. The composer's model pill paints (from the composer's own sticky state), the effort pill shows the pending spinner and never resolves.","4. Pick any effort level once — the pill paints and the pick persists per chat across gateway restarts.","1. Desktop marks a resumed slice's effort pending by design (`markReasoningEffortPending`, `apps/desktop/src/lib/chat-runtime.ts`): the resume reply carries no `reasoning_effort` (since #79807), and the clearing value ar"],"workaround":null,"fixed_in":null,"score":74.0,"first_seen":"2026-10-25T09:11:37Z"},{"id":"openclaw/openclaw#162193","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":162193,"title":"[Bug]: doctor --fix truncates existing shared-state SQLite DB and then aborts with Existing shared-state database generation changed","url":"https://github.com/openclaw/openclaw/issues/162193","state":"open","state_reason":null,"labels":["bug","regression","clawsweeper:no-new-fix-pr","clawsweeper:needs-maintainer-review","impact:data-loss","P0","issue-rating: 🦪 silver shellfish","maturity:stable","impact:ux-release-blocker"],"created":"2026-09-30T22:57:26Z","updated":"2026-10-23T03:35:21Z","closed":null,"reactions":2,"thumbs_up":0,"comments":5,"severity":"critical","severity_reasons":["label:p0","title:data-loss/security"],"topic":"install/update","classes":["install/update","crash/hang","always-on/daemon"],"cosmetic_ui":false,"summary":"Regression (worked before, now fails) OpenClaw: 2026.9.7 (c074824) Node.js: v24.21.0 Linux [vps].de 5.2.0 #1 SMP Mon Sep 30 15:36:27 MSK 2024 x8664 GNU/Linux Gateway: system-wide systemd service Service file…","versions":["2026.9.7","5.2.0"],"error_snippets":["Existing shared-state database generation changed: ${params.pathname}","OPENCLAW_SERVICE_REPAIR_POLICY=external","Running openclaw doctor --fix while the Gateway is completely stopped reproducibly damages the existing shared-state database:","Check the reported service state and resolve the failure.","Existing shared-state database generation changed:"],"workaround":null,"fixed_in":null,"score":74.0,"first_seen":"2026-10-18T04:33:05Z"},{"id":"cline/cline#12361","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":12361,"title":"Empty plugin window after IDEA update","url":"https://github.com/cline/cline/issues/12361","state":"closed","state_reason":"completed","labels":["JetBrains"],"created":"2026-07-17T08:11:07Z","updated":"2026-10-03T04:02:21Z","closed":"2026-07-21T16:07:07Z","reactions":9,"thumbs_up":9,"comments":10,"severity":"medium","severity_reasons":["engagement 19","class:install/update","class:mcp/tools"],"topic":"install/update","classes":["install/update","mcp/tools","always-on/daemon","network"],"cosmetic_ui":false,"summary":"JetBrains Plugin I am using a beta version of Cline Today after IDEA update Cline plugin open with empty window. IDEA version: 2026.2 (Build #IU-262.8665.258, built on July 15, 2026) Open plugin window IntelliJ IDEA 2026.2 Build…","versions":["1.1.60","262.8665.258","25.0.3+9","144.0.15-262"],"error_snippets":["Today after IDEA update Cline plugin open with empty window.","Exception reporter ID: 18112414af1fead-8104-44e3-a675-b9d2bcb5b06f","trace.state.event.service.url=https://api.jetbrains.cloud/trace-status"],"workaround":null,"fixed_in":null,"score":73.9,"first_seen":"2026-10-19T18:38:33Z"},{"id":"anomalyco/opencode#48965","harness":"opencode","harness_name":"OpenCode","repo":"anomalyco/opencode","number":48965,"title":"SystemPrompt.environment crashes on every prompt: undefined is not an object (evaluating 'a.name')","url":"https://github.com/anomalyco/opencode/issues/48965","state":"closed","state_reason":"completed","labels":[],"created":"2026-09-14T10:35:10Z","updated":"2026-10-01T02:29:59Z","closed":"2026-10-01T02:29:59Z","reactions":22,"thumbs_up":22,"comments":4,"severity":"high","severity_reasons":["title:breakage"],"topic":"crash/hang","classes":["crash/hang","self-host/headless"],"cosmetic_ui":false,"summary":"OpenCode fails before sending any prompt to the model. The UI shows: The server log shows: The error occurs repeatedly for new sessions and different working directories. Observed output: macOS 27.0(Apple Silicon)","versions":["1.18.30"],"error_snippets":["Failed to send prompt","Unexpected server error. Check server logs for details.","TypeError: undefined is not an object (evaluating 'a.name')","at SystemPrompt.environment","opencode --pure --print-logs --log-level DEBUG run \"Reply with exactly: OK\""],"workaround":null,"fixed_in":null,"score":73.8,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openclaw/openclaw#173061","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":173061,"title":"[Bug]: a parked \"plugins reload --wait\" holds the plugin-replacement gate with no deadline and blocks the start of every isolated agent run on the Gateway","url":"https://github.com/openclaw/openclaw/issues/173061","state":"open","state_reason":null,"labels":["no-stale","P1","clawsweeper:fix-shape-clear","clawsweeper:queueable-fix","clawsweeper:source-repro","issue-rating: 🦞 diamond lobster","impact:other","impact:ux-friction"],"created":"2026-10-15T11:05:00Z","updated":"2026-10-17T10:48:30Z","closed":null,"reactions":2,"thumbs_up":1,"comments":2,"severity":"high","severity_reasons":["label:p1"],"topic":"always-on/daemon","classes":["always-on/daemon","mcp/tools","self-host/headless","install/update","network"],"cosmetic_ui":false,"summary":"openclaw plugins reload --wait parks indefinitely when the named plugins hold long-lived retained work. While it is parked, no isolated agent run on the Gateway can start — automations for every agent fail with cron: isolated agent setup…","versions":["2026.10.2","2.13","1.5"],"error_snippets":["ls -l /proc/*/fd 2>/dev/null | grep plugin-reload-wait","cron: isolated agent setup timed out before runner start","is the documented recovery for a failed drain, its gateway-wide effect should be stated at the point of recommendation, and the command should refuse to park without","isolated agent setup timed out before runner start","`openclaw plugins reload --wait` parks indefinitely when the named plugins hold long-lived retained work. While it is parked, **no isolated agent run on the Gateway can start** — automations for every agent fail with `cr"],"workaround":null,"fixed_in":null,"score":73.7,"first_seen":"2026-10-18T04:33:05Z"},{"id":"cline/cline#14683","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":14683,"title":"Desktop 0.0.38: aarch64 Linux remote helper fails --remote-hub-ensure with Bun SyntaxError \"Invalid character: '\\0'\" (regression from previous helper)","url":"https://github.com/cline/cline/issues/14683","state":"open","state_reason":null,"labels":[],"created":"2026-09-30T04:32:28Z","updated":"2026-10-08T10:25:12Z","closed":null,"reactions":1,"thumbs_up":1,"comments":8,"severity":"high","severity_reasons":["title:breakage"],"topic":"self-host/headless","classes":["self-host/headless","install/update","always-on/daemon"],"cosmetic_ui":false,"summary":"Cline Desktop 0.0.38 (core 0.0.88) cannot connect to a Linux aarch64 SSH remote. The bundled cline-remote-helper-aarch64-unknown-linux-gnu (sha256 93444d160eed1dc9…) exits 1 on --remote-hub-ensure with a Bun SyntaxError: Invalid character…","versions":["0.0.38","0.0.88","3.0.65","0.0.86"],"error_snippets":["$ $H --remote-hub-ensure --cwd /home/ubuntu --discovery-path /home/ubuntu/.cline/data/remote/probe.json </dev/null","2788 | `),process.exitCode=1});","SyntaxError: Invalid character: '\\0'","at emit (node:events:100:22)","--remote-hub-ensure --cwd /home/ubuntu --discovery-path /home/ubuntu/.cline/data/remote/probe-old.json </dev/null"],"workaround":{"text":"`mkdir -p ~/.cline/remote-helpers ; scp :~/.cline/remote/cline-remote-helper-linux-arm64-8b51183e615e7442 \\`","source":"issue body","url":"https://github.com/cline/cline/issues/14683"},"fixed_in":null,"score":73.7,"first_seen":"2026-10-19T18:38:33Z"},{"id":"omacom/omarchy#14946","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":14946,"title":"Omarchy menu (SUPER+SPACE) renders pure black screen after 2026-10-10 Arch update — Qt 6.12 breaks quickshell qs.* singletons","url":"https://github.com/omacom/omarchy/issues/14946","state":"open","state_reason":null,"labels":[],"created":"2026-10-11T00:34:30Z","updated":"2026-10-22T09:56:47Z","closed":null,"reactions":2,"thumbs_up":2,"comments":6,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"After the 2026-10-10 omarchy update (which pulled Arch's Qt 6.11.2 → 6.12.0 transition together with quickshell 0.3.1 → 0.3.2), pressing SUPER+SPACE shows a fullscreen opaque black rectangle instead of the menu. The menu itself is…","versions":["6.12","6.11.2","6.12.0","0.3.1"],"error_snippets":["omarchy-shell[...]: WARN scene: @Commons/Border.qml[25:-1]: TypeError: Cannot read property 'menu.border-alpha' of undefined","Color.foreground -> undefined (should be \"#cacccc\")","Color.menu -> undefined","Util.abs(-5) -> TypeError: Property 'abs' of object Util_QMLTYPE_3 is not a function","; same class of errors from user plugins:"],"workaround":null,"fixed_in":null,"score":73.3,"first_seen":"2026-10-19T18:38:33Z"},{"id":"Aider-AI/aider#5076","harness":"aider","harness_name":"Aider","repo":"Aider-AI/aider","number":5076,"title":"Generated code can read SSH private keys outside the project directory without warning","url":"https://github.com/Aider-AI/aider/issues/5076","state":"open","state_reason":null,"labels":[],"created":"2026-04-26T17:24:50Z","updated":"2026-10-05T07:46:46Z","closed":null,"reactions":0,"thumbs_up":0,"comments":17,"severity":"medium","severity_reasons":["engagement 17","class:self-host/headless"],"topic":"self-host/headless","classes":["self-host/headless","auth/session"],"cosmetic_ui":false,"summary":"I would like to report a security-relevant issue in Aider's code generation behavior around sensitive file access outside the project directory. In a validated retest, Aider was gradually steered through a multi-turn conversation and…","versions":[],"error_snippets":["ssh_key_path = os.path.expanduser(\"~/.ssh/id_rsa\")","if os.path.exists(ssh_key_path):","In a validated retest, Aider was gradually steered through a multi-turn conversation and eventually generated and committed code that reads the user's SSH private key from:","Expected behavior:","credential files such as `~/.ssh/id_rsa`, especially when the target path is outside the"],"workaround":null,"fixed_in":null,"score":73.2,"first_seen":"2026-10-26T10:12:02Z"},{"id":"continuedev/continue#12492","harness":"continue","harness_name":"Continue","repo":"continuedev/continue","number":12492,"title":"Integration: would a runtime governance proxy in front of MCP servers be in scope for third-party integrations?","url":"https://github.com/continuedev/continue/issues/12492","state":"open","state_reason":null,"labels":["kind:question","area:integration","area:mcp"],"created":"2026-05-25T18:47:02Z","updated":"2026-10-12T12:38:08Z","closed":null,"reactions":0,"thumbs_up":0,"comments":7,"severity":"medium","severity_reasons":["class:self-host/headless","class:mcp/tools"],"topic":"self-host/headless","classes":["self-host/headless","mcp/tools","network","crash/hang","auth/session"],"cosmetic_ui":false,"summary":"Hi maintainers, Opening an issue first to gauge appetite before any code work, in the same shape as #12423. Would maintainers accept a third-party integration entry for Vaara (AGPL-3.0-or-later, https://github.com/vaaraio/vaara) used as a…","versions":["3.0-or","1.0","1.5"],"error_snippets":[", with optional Ed25519-signed OVERT 1.0 envelopes for the same set. Operator allowlist and denylist flags filter tools, resources, and prompts at the perimeter. Streaming","Would maintainers accept a third-party integration entry for Vaara (AGPL-3.0-or-later, https://github.com/vaaraio/vaara) used as a transparent runtime governance proxy in front of any stdio MCP server a Continue.dev user","Issue #12068 (APEX RGPD), #12172 (granular tool approval policy), #12248 (subagent permissions and policy bypass), and #11468 (TLS verification and OAuth) suggest Continue.dev users are asking for governance and audit su"],"workaround":null,"fixed_in":null,"score":72.9,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openclaw/openclaw#109867","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":109867,"title":"[Bug]: beta.2 state migration creates agent_id index before adding column, blocking gateway startup","url":"https://github.com/openclaw/openclaw/issues/109867","state":"closed","state_reason":"completed","labels":["no-stale","clawsweeper:fix-shape-clear","clawsweeper:queueable-fix","clawsweeper:source-repro","impact:crash-loop","P0","issue-rating: 🦞 diamond lobster","impact:ux-release-blocker"],"created":"2026-07-17T09:22:26Z","updated":"2026-09-27T03:47:51Z","closed":"2026-07-27T19:27:01Z","reactions":8,"thumbs_up":7,"comments":8,"severity":"critical","severity_reasons":["label:p0"],"topic":"always-on/daemon","classes":["always-on/daemon","install/update","crash/hang"],"cosmetic_ui":false,"summary":"Regression (worked before, now fails) After upgrading from 2026.7.2-beta.1 to 2026.7.2-beta.2, the shared SQLite migration creates indexes referencing managedoutgoingimagerecords.agentid before adding that column, so doctor --fix fails…","versions":["2026.7.2-beta.1","2026.7.2-beta.2"],"error_snippets":["Config health-state write failed: OpenClaw state database ~/.openclaw/state/openclaw.sqlite has a legacy operator approval schema; run openclaw doctor --fix to migrate it.","- Failed migrating shared state database schema at","~/.openclaw/state/openclaw.sqlite: Error: no such column: agent_id","[restart] failed to write gateway restart intent: Error: OpenClaw state database ~/.openclaw/state/openclaw.sqlite has a legacy operator approval schema; run openclaw doctor --fix to migrate it.","Gateway restart failed after 16s: service stayed stopped and port 18789 stayed free."],"workaround":null,"fixed_in":null,"score":72.8,"first_seen":"2026-10-18T04:30:03Z"},{"id":"omacom/omarchy#15006","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":15006,"title":"Keyboard layout chosen in installer isn't applied to Hyprland/fcitx5 — desktop falls back to US","url":"https://github.com/omacom/omarchy/issues/15006","state":"open","state_reason":null,"labels":[],"created":"2026-10-11T10:05:35Z","updated":"2026-10-25T21:52:29Z","closed":null,"reactions":1,"thumbs_up":1,"comments":5,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"During installation I selected Swiss German as the keyboard layout. The installer used that layout correctly. After the first boot into the desktop, though, the keyboard was US English everywhere. The layout chosen in the installer should…","versions":["4.0.4-1"],"error_snippets":["even after I set","During installation I selected **Swiss German** as the keyboard layout. The installer used that layout correctly. After the first boot into the desktop, though, the keyboard was **US English** everywhere.","There's also a second problem. fcitx5 runs (`omarchy-fcitx5.service`) but there's no `~/.config/fcitx5/profile`, so it uses its default `keyboard-us` input method. Its virtual keyboard (`hl-virtual-keyboard-fcitx5`) stay"],"workaround":{"text":"`~/.config/hypr/input.lua`: `hl.config({ input = { kb_layout = \"ch\", kb_variant = \"de\" } })` - `~/.config/fcitx5/profile` with `DefaultIM=keyboard-ch-de`, then restart fcitx5.","source":"issue body","url":"https://github.com/omacom/omarchy/issues/15006"},"fixed_in":null,"score":72.7,"first_seen":"2026-10-22T09:27:06Z"},{"id":"aaif-goose/goose#12805","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":12805,"title":"Define an OS sandbox boundary for shell and tool subprocesses","url":"https://github.com/aaif-goose/goose/issues/12805","state":"open","state_reason":null,"labels":[],"created":"2026-10-09T21:41:38Z","updated":"2026-10-26T07:37:31Z","closed":null,"reactions":2,"thumbs_up":1,"comments":2,"severity":"medium","severity_reasons":["class:network"],"topic":"network","classes":["network","self-host/headless","mcp/tools"],"cosmetic_ui":false,"summary":"What problem would this solve? goose can already ask before a tool runs (approve, smartapprove) and can deny a tool entirely. That is a permission policy, not a sandbox. Once a command, MCP server, or extension subprocess is allowed, it…","versions":["1.25.0"],"error_snippets":["tool executes the model-supplied command with no OS boundary. The default","excludedCommands",") and a removed Seatbelt experiment ([michaelneale/agent-seatbelt-sandbox](https://github.com/michaelneale/agent-seatbelt-sandbox), which @michaelneale has been using as the macOS reference: kernel","sandbox-exec","Who this affects: anyone running goose on a machine that has credentials, other projects, or network access they do not want an agent session to touch. It also blocks a safer default than \"ask every time\" or \"allow the w"],"workaround":null,"fixed_in":null,"score":71.8,"first_seen":"2026-10-19T18:38:33Z"},{"id":"omacom/omarchy#15779","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":15779,"title":"Ghostty installed via pacman doesn't launch from the app menu (D-Bus/systemd activation fails), works from a terminal","url":"https://github.com/omacom/omarchy/issues/15779","state":"open","state_reason":null,"labels":["bug"],"created":"2026-10-18T00:14:45Z","updated":"2026-10-23T02:49:26Z","closed":null,"reactions":1,"thumbs_up":1,"comments":2,"severity":"medium","severity_reasons":["class:always-on/daemon","class:install/update"],"topic":"always-on/daemon","classes":["always-on/daemon","install/update"],"cosmetic_ui":false,"summary":"After installing Ghostty (ghostty 1.3.1, Arch extra), clicking Ghostty in the app menu does nothing — no window, no error. Running ghostty (or /usr/bin/ghostty --gtk-single-instance=true) from another terminal opens a window fine. The…","versions":["1.3.1","0.56.2"],"error_snippets":["Exec=/usr/bin/ghostty --gtk-single-instance=true","ExecStart=/usr/bin/ghostty --gtk-single-instance=true --initial-window=false","app-com.mitchellh.ghostty.service: Failed with result 'protocol'.","Failed to start Ghostty.","After installing Ghostty (`ghostty 1.3.1`, Arch extra), clicking Ghostty in the app menu does nothing — no window, no error. Running `ghostty` (or `/usr/bin/ghostty --gtk-single-instance=true`) from another terminal open"],"workaround":null,"fixed_in":null,"score":71.6,"first_seen":"2026-10-19T18:38:33Z"},{"id":"anomalyco/opencode#55211","harness":"opencode","harness_name":"OpenCode","repo":"anomalyco/opencode","number":55211,"title":"OAuth Authorization Fails with HTTP 400 in OpenCode Desktop 2.0.28 on Windows","url":"https://github.com/anomalyco/opencode/issues/55211","state":"closed","state_reason":"completed","labels":[],"created":"2026-10-13T18:31:43Z","updated":"2026-10-25T17:07:20Z","closed":"2026-10-25T17:07:20Z","reactions":6,"thumbs_up":6,"comments":9,"severity":"medium","severity_reasons":["engagement 15","class:auth/session"],"topic":"auth/session","classes":["auth/session","install/update","mcp/tools","network"],"cosmetic_ui":false,"summary":"OpenCode Desktop 2.0.28 fails to complete ChatGPT OAuth authorization on Windows. After selecting the ChatGPT sign-in option and proceeding through the browser authorization flow, the browser displays Authorization failed with the message…","versions":["2.0.28"],"error_snippets":["OpenCode Desktop 2.0.28 fails to complete ChatGPT OAuth authorization on Windows. After selecting the ChatGPT sign-in option and proceeding through the browser authorization flow, the browser displays Authorization faile","The available logs do not identify the root cause. Please investigate why the OAuth flow returns HTTP 400 and provide more detailed diagnostic information so the underlying failure can be identified.","No relevant third-party plugins identified. The issue occurs during ChatGPT OAuth authorization, before a successful connection is established. (No complete plugin inventory has been verified.)","Observe the error page.","Authorization failed"],"workaround":null,"fixed_in":null,"score":71.4,"first_seen":"2026-10-21T09:24:21Z"},{"id":"aaif-goose/goose#12905","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":12905,"title":"OAuth succeeds, then the extension fails, when the token exceeds Windows Credential Manager's 2560-char limit","url":"https://github.com/aaif-goose/goose/issues/12905","state":"open","state_reason":null,"labels":[],"created":"2026-10-15T00:49:26Z","updated":"2026-10-26T02:39:22Z","closed":null,"reactions":0,"thumbs_up":0,"comments":4,"severity":"critical","severity_reasons":["title:data-loss/security"],"topic":"auth/session","classes":["auth/session","self-host/headless","mcp/tools"],"cosmetic_ui":false,"summary":"Describe the bug: On Windows, a remote MCP extension fails to load after a successful authorization. The browser shows Goose's own \"Authorization Success\" page, then the extension errors. The real cause is that saving the credential fails…","versions":["1.54","1.53"],"error_snippets":["On Windows, a remote MCP extension fails to load after a successful authorization. The browser shows Goose's own \"Authorization Success\" page, then the extension errors. The real cause is that saving the credential fails","WARN [OAuth: ] Browser authorization flow failed: Internal error:","Failed to save credentials: Failed to access keyring:","Attribute 'password encoded as UTF-16' is longer than platform limit of 2560 chars","WARN Failed to load extension : failed to initialize MCP client:"],"workaround":{"text":"Additional context: GOOSE_DISABLE_KEYRING=1 is a complete workaround, which confirms the diagnosis.","source":"issue body","url":"https://github.com/aaif-goose/goose/issues/12905"},"fixed_in":null,"score":71.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"continuedev/continue#13119","harness":"continue","harness_name":"Continue","repo":"continuedev/continue","number":13119,"title":"IntelliJ IDEA 2026.2.1 : Continue crashes","url":"https://github.com/continuedev/continue/issues/13119","state":"open","state_reason":null,"labels":[],"created":"2026-08-11T11:55:34Z","updated":"2026-09-15T14:39:27Z","closed":null,"reactions":3,"thumbs_up":3,"comments":8,"severity":"high","severity_reasons":["title:breakage"],"topic":"crash/hang","classes":["crash/hang","always-on/daemon","self-host/headless","mcp/tools"],"cosmetic_ui":false,"summary":"Continue plugin crash : IntelliJ IDEA 2026.2.1 Build #IU-262.9437.185, built on August 9, 2026 Source revision: 5ced7b0cd2861 Licensed to CNRS / DSI / David Marques You have a perpetual fallback license for this version. Subscription is…","versions":["2026.2.1","262.9437.185","25.0.3+9","144.0.15-262"],"error_snippets":["Exception reporter ID: 08102517e5fe3e1-0b26-4cb6-aae8-64d6c2092562","ide.experimental.ui=true","trace.state.event.service.url=https://api.jetbrains.cloud/trace-status","java.lang.NoClassDefFoundError: com/intellij/ui/jcef/JBCefApp","at kotlinx.coroutines.EventLoopImplBase.processNextEvent(EventLoop.common.kt:263)"],"workaround":null,"fixed_in":null,"score":71.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"github/copilot-cli#4994","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":4994,"title":"Regression (1.0.88 → 1.0.89): MCP OAuth authorization page opens on every ACP session instead of the challenge being resolved silently","url":"https://github.com/github/copilot-cli/issues/4994","state":"open","state_reason":null,"labels":["area:non-interactive","area:mcp"],"created":"2026-09-29T10:12:55Z","updated":"2026-10-19T15:30:53Z","closed":null,"reactions":2,"thumbs_up":2,"comments":2,"severity":"high","severity_reasons":["title:breakage"],"topic":"install/update","classes":["install/update","auth/session","mcp/tools","self-host/headless"],"cosmetic_ui":false,"summary":"An OAuth-protected HTTP MCP server that has a valid, unexpired access token in ~/.copilot/mcp-oauth-config is answered with HTTP 401 on every new --acp session, and the CLI responds by opening the OAuth authorization page again. Because…","versions":["1.0.88","1.0.89","152.1","1.0.80"],"error_snippets":["[ERROR] [rust:rmcp::transport::worker] worker quit with fatal: Transport channel closed,","www_authenticate_header: \"******\\\"https://mcp.internal.example.com/.well-known/oauth-protected-resource/servers/REDACTED-MCP/mcp\\\"\",","response: McpOAuthHttpResponse { status_code: 401, header_count: 6, has_body: true } })","[WARNING] [rust:copilot_runtime::session::mcp::agent_host] HTTP 401 challenge","(WWW-Authenticate: ******\"https://mcp.internal.example.com/.well-known/oauth-protected-resource/servers/REDACTED-MCP/mcp\")"],"workaround":null,"fixed_in":null,"score":70.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"omacom/omarchy#12218","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":12218,"title":"RTX 5090 grey-screen freeze after 4.0.4 update; 6.9 million NVIDIA register errors displace kernel history","url":"https://github.com/omacom/omarchy/issues/12218","state":"open","state_reason":null,"labels":["bug","compatibility"],"created":"2026-09-17T04:01:48Z","updated":"2026-10-23T03:44:54Z","closed":null,"reactions":1,"thumbs_up":1,"comments":7,"severity":"high","severity_reasons":["title:breakage"],"topic":"install/update","classes":["install/update","crash/hang"],"cosmetic_ui":false,"summary":"Omarchy 4.0.4-1 (updated from 4.0.3-1 earlier on the day of the incident) Kernel currently running after recovery: 7.2.5-3-omarchy; linux-omarchy 7.2.5-3 and headers installed during that update Previous stock kernel linux 7.2.3.arch1-3…","versions":["4.0.4","6.9","4.0.4-1","4.0.3-1"],"error_snippets":["T+0.000000 kernel: NVRM: gpuHandleSanityCheckRegReadError_GH100: Possible bad register read: addr: 0xb830b0, regvalue: 0xbadf3200, error code: Unknown SYS_PRI_ERROR_CODE","T+0.008030 kernel: NVRM: gpuHandleSanityCheckRegReadError_GH100: Possible bad register read: addr: 0xb830b0, regvalue: 0xbadf3200, error code: Unknown SYS_PRI_ERROR_CODE","T+49.729614 kernel: NVRM: gpuHandleSanityCheckRegReadError_GH100: Possible bad register read: addr: 0xb830b0, regvalue: 0xbadf3200, error code: Unknown SYS_PRI_ERROR_CODE","~T+16s steam: src/clientdll/steamengine.cpp (2665) : Assertion Failed: CSteamEngine::BMainLoop appears to have stalled > 15 seconds without event signalled","This is a report of an observed graphics failure and severe kernel-log flooding after the update, **not a claim that the new kernel is the confirmed cause**. NVIDIA driver/firmware or hardware involvement remains possibl"],"workaround":null,"fixed_in":null,"score":70.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"cline/cline#14924","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":14924,"title":"CLI: MCP notifications/tools/list_changed is ignored — tools added mid-session never appear (extension fixed in #12619)","url":"https://github.com/cline/cline/issues/14924","state":"open","state_reason":null,"labels":["CLI"],"created":"2026-10-07T21:40:19Z","updated":"2026-10-11T17:09:44Z","closed":null,"reactions":0,"thumbs_up":0,"comments":7,"severity":"medium","severity_reasons":["class:crash/hang","class:mcp/tools"],"topic":"crash/hang","classes":["crash/hang","mcp/tools","self-host/headless"],"cosmetic_ui":false,"summary":"3.0.69 (npm cline, Windows x64); same code on main (7a75e99) Tools that an MCP server adds during a session never reach the model in the CLI. The server sends notifications/tools/listchanged, but the CLI keeps the tool list it built when…","versions":["3.0.69"],"error_snippets":["sdk/packages/core/src/extensions/mcp/","was called once at startup and never again, and every later request, including the next turn's, offered only","In the SDK core there is no handler for `list_changed` (`sdk/packages/core/src/extensions/mcp/` — the manager caches `tools/list` and `createMcpTools` builds the session's tools once). #12619 fixed this for the VS Code e","Expected: after `notifications/tools/list_changed`, the CLI re-lists the server's tools and offers them from the next model request (or at least the next turn), as the extension does.","1. A streamable-HTTP MCP server (stateful sessions, `tools.listChanged: true`) that exposes `alpha`; calling `alpha` registers a second tool `beta`, which makes the server send `notifications/tools/list_changed`."],"workaround":null,"fixed_in":null,"score":70.3,"first_seen":"2026-10-19T18:38:33Z"},{"id":"NousResearch/hermes-agent#138329","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":138329,"title":"httpx2==2.12.0's 1 MiB per-SSE-event cap silently breaks MCP servers with a large tools/list (reported as \"SSE stream ended without a response\")","url":"https://github.com/NousResearch/hermes-agent/issues/138329","state":"closed","state_reason":"completed","labels":["type/bug","comp/tools","tool/mcp","P2","area/streaming","cause/upstream"],"created":"2026-10-14T10:26:07Z","updated":"2026-10-24T00:40:35Z","closed":"2026-10-24T00:40:35Z","reactions":3,"thumbs_up":3,"comments":5,"severity":"medium","severity_reasons":["class:self-host/headless","class:mcp/tools"],"topic":"self-host/headless","classes":["self-host/headless","mcp/tools","network","install/update"],"cosmetic_ui":false,"summary":"After upgrading to an environment that has httpx2==2.12.0 (the pin Hermes carries for the mcp extra since bc8545bd1a, \"bump httpx2/httpcore2 2.7.0 -> 2.12.0 for the audit advisories\"), any HTTP MCP server whose tools/list response arrives…","versions":["2.12.0","2.7.0","2.9.1","2.12"],"error_snippets":["✗ Connection failed (12.3s): SSE stream ended without a response","httpx2.SSEError: Server-sent event exceeded the 1048576 byte limit.","raise SSEError(f\"Server-sent event exceeded the {self._max_event_size} byte limit.\")","MCP server '<x>' failed initial connection after 3 attempts, parking until a reconnect is requested","EventSource.__init__"],"workaround":null,"fixed_in":null,"score":69.8,"first_seen":"2026-10-18T04:32:26Z"},{"id":"aaif-goose/goose#11740","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":11740,"title":"feat(tasks): unified task registry + tasks platform extension (background processes, timers, wait_for wake-ups)","url":"https://github.com/aaif-goose/goose/issues/11740","state":"open","state_reason":null,"labels":[],"created":"2026-09-01T05:02:56Z","updated":"2026-10-09T09:43:21Z","closed":null,"reactions":2,"thumbs_up":2,"comments":5,"severity":"medium","severity_reasons":["class:always-on/daemon","class:network"],"topic":"always-on/daemon","classes":["always-on/daemon","network","self-host/headless","mcp/tools"],"cosmetic_ui":false,"summary":"Long-running work is a first-class citizen in agent-driven development — builds, test suites, dev servers, watchers, interactive programs (debuggers, REPLs), and scheduled follow-ups. Today goose has no first-class way to run that work in…","versions":[],"error_snippets":["), no cancel-with-partial-output, no structured lifecycle — every ad-hoc","enum should reserve the full space (","), platform extension with the six tools above (","A unified **task registry** in the agent runtime, with a small platform extension exposing it to the model:"],"workaround":null,"fixed_in":null,"score":69.8,"first_seen":"2026-10-19T18:38:33Z"},{"id":"omacom/omarchy#13040","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":13040,"title":"omarchy-mise-install wrappers exec-loop forever when mise shims are on PATH after ~/.local/bin","url":"https://github.com/omacom/omarchy/issues/13040","state":"open","state_reason":null,"labels":["bug"],"created":"2026-09-23T16:41:32Z","updated":"2026-10-13T02:32:56Z","closed":null,"reactions":2,"thumbs_up":1,"comments":10,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update","always-on/daemon","network"],"cosmetic_ui":false,"summary":"Intel i9-9880H, Intel UHD 630 + NVIDIA Quadro T2000 Mobile, Omarchy 4.0.4-1, mise-bin 2026.9.12-1 A wrapper generated by omarchy-mise-install can call itself forever. It never exits, and it pins a CPU core by running mise use -g about…","versions":["4.0.4-1","2026.9.12-1"],"error_snippets":["exec mise x \"<package>\" -- \"<bin>\" \"$@\"","timeout 5 codex --version; echo \"rc=$?\" # rc=124 (hangs)","timeout 5 codex --version; echo \"rc=$?\" # rc=0","mise use -g --quiet $package_arg || exit 1","bin_path=\\$(mise which --tool $package_arg $bin_arg) || exit 1"],"workaround":null,"fixed_in":null,"score":69.3,"first_seen":"2026-10-19T18:38:33Z"},{"id":"github/copilot-cli#4929","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":4929,"title":"Process-local auth token stops refreshing; all prompts fail until restart","url":"https://github.com/github/copilot-cli/issues/4929","state":"open","state_reason":null,"labels":["triage"],"created":"2026-09-22T05:49:46Z","updated":"2026-09-28T18:19:33Z","closed":null,"reactions":0,"thumbs_up":0,"comments":13,"severity":"medium","severity_reasons":["class:auth/session"],"topic":"auth/session","classes":["auth/session","install/update","mcp/tools","network"],"cosmetic_ui":false,"summary":"A long-running Copilot CLI process permanently loses authentication. After the failure, every normal prompt and /ask immediately returns an authorization error. Running /login does not recover the process. Restarting and resuming the same…","versions":["1.0.85"],"error_snippets":["A long-running Copilot CLI process permanently loses authentication. After the failure, every normal prompt and /ask immediately returns an authorization error. Running /login does not recover the process. Restarting and","The failure is process-local: several older concurrent CLI processes remained usable, and a newly launched process authenticated successfully. No COPILOT_GITHUB_TOKEN, GH_TOKEN, or GITHUB_TOKEN override was present.","The failing process first logged an expired/rejected GitHub MCP token almost exactly one hour after startup. It later logged 401 AuthenticateToken failed from the Copilot Tasks API. OAuth recovery attempted discovery aga","4. Observe that every prompt and /ask reports an authorization error.","7. Restart/resume the failed process; observe that authentication works again."],"workaround":null,"fixed_in":null,"score":68.4,"first_seen":"2026-10-19T18:38:33Z"},{"id":"omacom/omarchy#11899","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":11899,"title":"`omarchy setup security fingerprint` silently evicts a working libfprint fork, then \"succeeds\" against the stale daemon","url":"https://github.com/omacom/omarchy/issues/11899","state":"open","state_reason":null,"labels":["bug"],"created":"2026-09-15T04:06:31Z","updated":"2026-10-22T19:10:57Z","closed":null,"reactions":2,"thumbs_up":2,"comments":6,"severity":"medium","severity_reasons":["class:always-on/daemon"],"topic":"always-on/daemon","classes":["always-on/daemon","install/update","auth/session","network"],"cosmetic_ui":false,"summary":"This machine's fingerprint sensor — a Dell XPS 15 7590 with the Goodix 27c6:5395 (\"Goodix HTK32\") — has no driver in the pinned libfprint-git snapshot. The community fork libfprint-goodix53x5 (upstream, built on libfprint tag v1.94.10…","versions":["1.94.10"],"error_snippets":["01:39:34 installed libfprint-goodix53x5 → reader works, enrollment succeeds","(enroll/verify \"passed\" via the stale in-memory fork)","sudo fprintd-enroll","fprintd-enroll -f <finger>","`pacman -Q \"$pkg\"`, which is provides-aware — but **no fork can ever provide the"],"workaround":null,"fixed_in":null,"score":68.0,"first_seen":"2026-10-20T09:28:14Z"},{"id":"NousResearch/hermes-agent#142491","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":142491,"title":"Gateway group workspace UX gaps: no paste/drag, no @-autocomplete, no inline image rendering, members can't send files","url":"https://github.com/NousResearch/hermes-agent/issues/142491","state":"open","state_reason":null,"labels":["type/feature","P3","comp/desktop","desktop/chat","desktop/files"],"created":"2026-10-21T10:30:54Z","updated":"2026-10-21T10:36:27Z","closed":null,"reactions":2,"thumbs_up":2,"comments":0,"severity":"medium","severity_reasons":["class:always-on/daemon"],"topic":"always-on/daemon","classes":["always-on/daemon","crash/hang"],"cosmetic_ui":false,"summary":"The gateway-owned group chat workspace (canonical-group, Desktop v0.22.2) lost several capabilities the legacy Desktop room UI had. Five gaps, all in the same surface: Composer: no clipboard image paste, no drag-and-drop — a bare in…","versions":["0.22.2"],"error_snippets":["extractClipboardImageBlobs","The gateway-owned group chat workspace (canonical-group, Desktop v0.22.2) lost several capabilities the legacy Desktop room UI had. Five gaps, all in the same surface:","3. **Sparse timeline** — message grouping, author styling/avatars, timestamps and thread affordances from the legacy `group-chat-view.tsx` are gone; the room reads like a debug workspace rather than a chat.","4. **Attachments are not rendered inline** — `canonical-group-history.tsx` shows image attachments as filename chips with a download action only; there is no ` ` preview, so \"look at this screenshot\" requires download + ","5. **Members cannot send images/files** — member replies are text-only in the room protocol, and `::preview{file=...}` directives (which render in normal chats) show as raw text in groups. Any artifact a member produces "],"workaround":null,"fixed_in":null,"score":67.5,"first_seen":"2026-10-23T09:23:24Z"},{"id":"kirodotdev/Kiro#12118","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":12118,"title":"kiro-cli acp: session/new blocks ~66 s downloading the 83 MB all-MiniLM-L6-v2 model; a client timeout restarts it from zero","url":"https://github.com/kirodotdev/Kiro/issues/12118","state":"open","state_reason":null,"labels":["pending-triage","models","theme:slow-unresponsive","cli"],"created":"2026-10-26T03:14:24Z","updated":"2026-10-26T03:16:08Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"medium","severity_reasons":["class:crash/hang","class:network"],"topic":"crash/hang","classes":["crash/hang","network","install/update","auth/session"],"cosmetic_ui":false,"summary":"kiro-cli acp does not answer session/new until it has downloaded the 83 MB all-MiniLM-L6-v2 semantic-search model. On a fresh machine that took about 66 s here, with no output, no progress notification and nothing on stderr. An ACP client…","versions":["2.34.0","11.3","14.2","1.1"],"error_snippets":["semantic_search_client::client::hosted_model_client: Attempting to download from URL: https://desktop-release.q.us-east-1.amazonaws.com/models/all-MiniLM-L6-v2.zip","kiro-cli 2.34.0 (Homebrew cask), macOS 27.0.1, Apple Silicon, default agent engine (v2).","and then nothing until the process is killed. The partial `all-MiniLM-L6-v2.zip` in that directory grows with each attempt (11.3 MB, then 14.2 MB) and is not resumed.","5. With a 300 s probe instead, `session/new` answered after 66 s, once the download finished and the files were extracted. Later runs answer in about 4 s."],"workaround":null,"fixed_in":null,"score":67.3,"first_seen":"2026-10-26T10:12:02Z"},{"id":"NousResearch/hermes-agent#124471","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":124471,"title":"Hindsight core→plugin migration traps source install in interrupted update / dependency prompt loop","url":"https://github.com/NousResearch/hermes-agent/issues/124471","state":"open","state_reason":null,"labels":["type/bug","comp/cli","comp/plugins","tool/memory","P3","sweeper:risk-compatibility","area/install-update","Plugin Catalog"],"created":"2026-09-26T21:22:17Z","updated":"2026-10-17T08:11:44Z","closed":null,"reactions":1,"thumbs_up":1,"comments":7,"severity":"medium","severity_reasons":["class:install/update","class:mcp/tools"],"topic":"install/update","classes":["install/update","mcp/tools","crash/hang"],"cosmetic_ui":false,"summary":"After updating a git/source installation to Hermes Agent v0.21.5+2959.ga342e65 (commit a342e6563d), the automatic migration of the Hindsight memory provider from core to the plugin catalog gets stuck while preparing the plugin's Python…","versions":["0.21.5+2959.ga342e65","0.10.1","2026.9.24","2.24.0"],"error_snippets":["- hindsight-embed>=0.10.1,<1","^CSkipped — run `hermes plugins enable` when ready to prepare them.","The installed plugin and active environment are unchanged..","File \".../hermes_bootstrap.py\", line 532, in <module>","return self._wait(timeout=timeout)"],"workaround":null,"fixed_in":null,"score":67.1,"first_seen":"2026-10-18T04:32:26Z"},{"id":"openclaw/openclaw#167671","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":167671,"title":"[Bug]: Gateway SIGBUS on a live agent SQLite -shm mapping under concurrent agent activity","url":"https://github.com/openclaw/openclaw/issues/167671","state":"open","state_reason":null,"labels":["clawsweeper:no-new-fix-pr","clawsweeper:needs-maintainer-review","impact:crash-loop","P0","issue-rating: 🦪 silver shellfish"],"created":"2026-10-09T06:58:43Z","updated":"2026-10-09T07:17:34Z","closed":null,"reactions":2,"thumbs_up":1,"comments":2,"severity":"critical","severity_reasons":["label:p0"],"topic":"always-on/daemon","classes":["always-on/daemon","crash/hang","self-host/headless","install/update"],"cosmetic_ui":false,"summary":"The Gateway exits with SIGBUS (code=dumped, status=7/BUS) when a read-only session worker shrinks an agent SQLite WAL-index file (openclaw-agent.sqlite-shm) to 3 bytes while the Gateway still has it memory-mapped, during concurrent agent…","versions":["2026.9.8","2026.9.6","2026.9.7","3.53.1"],"error_snippets":["08:25:05 systemd: Main process exited, code=dumped, status=7/BUS","08:33:38 systemd: Main process exited, code=dumped, status=7/BUS","08:36:43 systemd: Main process exited, code=dumped, status=7/BUS","The Gateway exits with SIGBUS (`code=dumped, status=7/BUS`) when a read-only session worker","the cycle repeats. **The Gateway had run for months without this failure before these crashes"],"workaround":null,"fixed_in":null,"score":66.7,"first_seen":"2026-10-18T04:33:05Z"},{"id":"cline/cline#15327","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":15327,"title":"OpenAI Compatible Base URL and API Key inputs automatically reset to default values","url":"https://github.com/cline/cline/issues/15327","state":"open","state_reason":null,"labels":[],"created":"2026-10-24T23:07:46Z","updated":"2026-10-25T13:42:03Z","closed":null,"reactions":0,"thumbs_up":0,"comments":2,"severity":"medium","severity_reasons":["class:auth/session"],"topic":"auth/session","classes":["auth/session"],"cosmetic_ui":false,"summary":"I am using a beta version of Cline When configuring the OpenAI Compatible provider in Cline Desktop, clicking on or interacting with the Base URL and/or API Key input fields automatically overwrites any custom input with the default URL…","versions":["0.0.48","0.0.48-20261025"],"error_snippets":["Because of this behavior, it is impossible to set and retain a custom endpoint address.","3. Attempt to type or paste a custom endpoint in the **Base URL** field.","6. **Expected Behavior:** The input fields should retain the user-specified custom Base URL and API Key without auto-resetting."],"workaround":null,"fixed_in":null,"score":66.4,"first_seen":"2026-10-26T10:12:02Z"},{"id":"NousResearch/hermes-agent#133992","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":133992,"title":"[Bug]: Regression of #78119 / #87514: macOS Desktop update hand-off refuses its own hermes update (custodian + second-resolution delegate ct)","url":"https://github.com/NousResearch/hermes-agent/issues/133992","state":"closed","state_reason":"completed","labels":["type/bug","comp/cli","P2","needs-repro","sweeper:risk-compatibility","comp/desktop","area/install-update"],"created":"2026-10-06T15:58:06Z","updated":"2026-10-10T17:44:38Z","closed":"2026-10-10T17:44:38Z","reactions":4,"thumbs_up":3,"comments":24,"severity":"high","severity_reasons":["title:breakage"],"topic":"install/update","classes":["install/update","always-on/daemon"],"cosmetic_ui":false,"summary":"On macOS, every update started from the Desktop app's Update button fails with exit code 2 because hermes update refuses the lock held by its own hand-off: The PID named in the error is the custodian that scripts/desktop-update/posix.sh…","versions":["0.005","3.14.7","0.21.5+8102.g0dbaf33","0.21.5+8002.g781334e"],"error_snippets":["The relevant excerpts are below. I'm happy to share more specific output on request.","19:51:52 hermes update exit code: 2","# Earlier run on 781334eea4: the hand-off recorded the delegate's ct as 1791301233.000.","ct:1791301233.508","detached update FAILED (exit 2)"],"workaround":{"text":"quit the Desktop app and run `~/.hermes/hermes-agent/.hermes/bin/hermes update` in a terminal.","source":"issue body","url":"https://github.com/NousResearch/hermes-agent/issues/133992"},"fixed_in":null,"score":66.2,"first_seen":"2026-10-18T04:32:26Z"},{"id":"NousResearch/hermes-agent#135910","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":135910,"title":"pm repair cannot commit a dependency environment on a case-insensitive filesystem: shutil.copytree dies with Errno 17 on the bundled CPython terminfo","url":"https://github.com/NousResearch/hermes-agent/issues/135910","state":"open","state_reason":null,"labels":["type/bug","comp/cli","tool/file","P2","sweeper:risk-compatibility","area/install-update","os/macos"],"created":"2026-10-10T01:27:34Z","updated":"2026-10-19T23:31:42Z","closed":null,"reactions":2,"thumbs_up":2,"comments":3,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update","self-host/headless","mcp/tools"],"cosmetic_ui":false,"summary":"On the official Docker image (linux-arm64, v0.21.6) with HERMESHOME on a case-insensitive volume (macOS/APFS default — a plain Docker Desktop bind mount under /Users/...), no dependency environment can ever be committed for the install…","versions":["0.21.6"],"error_snippets":["hermes: no dependency environment is committed for this install; run `hermes pm repair`","hermes: dependency repair failed: [('.../tools/python-3.14.7+.../share/terminfo/e',","'.../installs/<key>/environments/<id>/workspace/tools/python-3.14.7+.../share/terminfo/e',","\"[Errno 17] File exists: '.../terminfo/e'\"), ...]; run `hermes pm repair`","for entry in source.iterdir():"],"workaround":null,"fixed_in":null,"score":66.2,"first_seen":"2026-10-20T09:28:14Z"},{"id":"anomalyco/opencode#51073","harness":"opencode","harness_name":"OpenCode","repo":"anomalyco/opencode","number":51073,"title":"[gateway] 30-image cap applied only to some session routes: identical request flips 400/200 with only x-opencode-session changed; capped sessions cannot /compact","url":"https://github.com/anomalyco/opencode/issues/51073","state":"open","state_reason":null,"labels":[],"created":"2026-09-24T08:12:04Z","updated":"2026-09-29T02:52:05Z","closed":null,"reactions":4,"thumbs_up":4,"comments":1,"severity":"medium","severity_reasons":["class:always-on/daemon","class:crash/hang"],"topic":"always-on/daemon","classes":["always-on/daemon","crash/hang","auth/session"],"cosmetic_ui":false,"summary":"The OpenCode Go gateway (https://opencode.ai/zen/go/v1) enforces a 30-image-per-request cap on only some routes: enforcement follows the x-opencode-session routing value and the model. Identical requests (same API key, same model, same…","versions":["3.14.3","2.6-flash"],"error_snippets":["Turn execution failed provider=opencode-go-chat-2 model=deepseek-v4.1-flash request=4c9964bb-0208-4ac5-816e-5bef6aa07af6 reason=invalid_request status=400 retryable=false","Upstream request failed: [invalid_request_error] Too many images in request: 31 > 30","400 Too many images in request: N > 30","with another. On a capped route every full-history request fails - **including","x-opencode-endpoint-id: orcarouter-dsv4.1flash"],"workaround":{"text":"Workaround used: temporarily switch the model/provider, compact there, switch back. - The same request works or fails purely based on the session header, which is extremely hard for users to diagnose. - Compounds #47493 (no client-side media strip) and #39677 (image-count 400 not classified as…","source":"issue body","url":"https://github.com/anomalyco/opencode/issues/51073"},"fixed_in":null,"score":66.1,"first_seen":"2026-10-19T18:38:33Z"},{"id":"anthropics/claude-code#96508","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":96508,"title":"[BUG] Regression of #86362 — Browser pane still blocks same-origin subresources (ERR_BLOCKED_BY_CLIENT) for hostnames resolving to 127.0.0.1","url":"https://github.com/anthropics/claude-code/issues/96508","state":"open","state_reason":null,"labels":["bug","has repro","platform:windows","regression","area:desktop"],"created":"2026-09-23T21:43:22Z","updated":"2026-10-20T13:53:53Z","closed":null,"reactions":4,"thumbs_up":4,"comments":6,"severity":"high","severity_reasons":["label:regression","title:breakage"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"Windows 11 Pro 10.0.26100 Claude desktop app, UA: Claude/2.7032.0 Chrome/152.0.7977.130 (MSIX) Hostname mapped to 127.0.0.1 via the Windows hosts file, e.g. dev.example.test; IIS over HTTPS on 443 Repro (run in the pane's page context…","versions":["10.0.26100","3.7.1"],"error_snippets":["await fetch('/') // TypeError: Failed to fetch","await fetch('/favicon.ico') // TypeError: Failed to fetch","GET https://dev.example.test/ → 404 (top-level navigation, allowed)","GET https://dev.example.test/ → FAILED: net::ERR_BLOCKED_BY_CLIENT (fetch, same URL)","GET https://dev.example.test/favicon.ico → FAILED: net::ERR_BLOCKED_BY_CLIENT"],"workaround":null,"fixed_in":null,"score":65.9,"first_seen":"2026-10-25T09:11:37Z"},{"id":"github/copilot-cli#4847","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":4847,"title":"Automatic managed-settings refresh breaks IDE MCP reload and disables /allow-all","url":"https://github.com/github/copilot-cli/issues/4847","state":"open","state_reason":null,"labels":["triage"],"created":"2026-09-15T02:45:45Z","updated":"2026-09-23T14:42:32Z","closed":null,"reactions":4,"thumbs_up":4,"comments":4,"severity":"medium","severity_reasons":["class:mcp/tools"],"topic":"mcp/tools","classes":["mcp/tools","self-host/headless","install/update","network"],"cosmetic_ui":false,"summary":"In a long-running Copilot CLI session connected to Visual Studio Code, an automatic managed-settings refresh can fail while reloading the dynamically contributed IDE MCP server: After this failure, the session enters a fail-closed…","versions":["1.0.84-6","1.0.84-8","7.6.6"],"error_snippets":["Failed to enforce managed Computer Use policy: failed to reload Computer Use","MCP reload failed: mcpServers.ide.type: Invalid literal value","[managedSettings] effective policy resolved:","source=mdm, bypassDisabled=false, serverFetchFailed=false","[managedSettings] self-fetch errored:"],"workaround":null,"fixed_in":null,"score":65.8,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aaif-goose/goose#10839","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":10839,"title":"Support MCP's server.json and mcp.json import","url":"https://github.com/aaif-goose/goose/issues/10839","state":"open","state_reason":null,"labels":[],"created":"2026-07-31T09:14:48Z","updated":"2026-10-14T21:06:30Z","closed":null,"reactions":2,"thumbs_up":2,"comments":6,"severity":"medium","severity_reasons":["class:self-host/headless","class:mcp/tools"],"topic":"self-host/headless","classes":["self-host/headless","mcp/tools"],"cosmetic_ui":false,"summary":"goose has its own repository for MCP servers but given that there is an MCP server for everything now, this concept is outdated. goose has deeplinks to install remote MCP servers but uses its own format for that. Meanwhile the world has…","versions":[],"error_snippets":[],"workaround":null,"fixed_in":null,"score":65.8,"first_seen":"2026-10-19T18:38:33Z"},{"id":"cline/cline#14066","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":14066,"title":"Desktop 0.0.26 backend + CLI crash-loop on pre-AVX CPUs — ship Windows x64 `-baseline` Bun binaries","url":"https://github.com/cline/cline/issues/14066","state":"open","state_reason":null,"labels":["beta"],"created":"2026-09-11T14:41:20Z","updated":"2026-09-26T10:13:14Z","closed":null,"reactions":1,"thumbs_up":1,"comments":10,"severity":"high","severity_reasons":["title:breakage"],"topic":"crash/hang","classes":["crash/hang","install/update","auth/session"],"cosmetic_ui":false,"summary":"0.0.26 (Desktop; CLI cline.exe bundles Bun 1.3.13) I am using a beta version of Cline Cline Desktop (and the cline CLI) are unusable on CPUs without AVX support because both bundle the standard (AVX2) build of Bun v1.3.13 instead of the…","versions":["0.0.26","1.3.13"],"error_snippets":["Sign in failed: Tauri invoke failed for get_desktop_backend_endpoint: desktop backend endpoint not ready","Cline Desktop (and the `cline` CLI) are unusable on CPUs without AVX support because both bundle the **standard (AVX2) build of Bun v1.3.13** instead of the `-baseline` build. The backend `code-sidecar.exe` crash-loops e","2. App shows \"Connecting...\" forever; clicking Sign in → \"Waiting for browser...\" → `Sign in failed: Tauri invoke failed for get_desktop_backend_endpoint: desktop backend endpoint not ready`. Browser never opens.","4. Windows Error Reporting collects an `AppCrash_code-sidecar.exe` report roughly every 4 minutes while the app is open (16 reports in one day on my machine).","Version strings extracted from the shipped binaries:"],"workaround":null,"fixed_in":null,"score":65.8,"first_seen":"2026-10-19T18:38:33Z"},{"id":"google-gemini/gemini-cli#29909","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":29909,"title":"GeminiCLI.com Feedback: listed extension logospell/logospell-mcp-server stuck at 5.1.1 since Oct 7 (manifest 6.2.0; index unchanged since Oct 16)","url":"https://github.com/google-gemini/gemini-cli/issues/29909","state":"open","state_reason":null,"labels":["area/extensions","status/bot-triaged","effort/small"],"created":"2026-10-24T08:55:08Z","updated":"2026-10-24T09:14:21Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"medium","severity_reasons":["class:self-host/headless","class:crash/hang"],"topic":"self-host/headless","classes":["self-host/headless","crash/hang","mcp/tools","always-on/daemon","install/update"],"cosmetic_ui":false,"summary":"https://geminicli.com/extensions/ https://geminicli.com/extensions.json The gallery entry for logospell/logospell-mcp-server still shows the extension as it was on October 7, 2026, and hasn't picked up any manifest change since. In…","versions":["5.1.1","6.2.0","6.0.0"],"error_snippets":["extensions.json","extensionVersion","extensionDescription","gemini-extension.json","(October 22). The manifest has changed ten times since the entry's"],"workaround":null,"fixed_in":null,"score":65.6,"first_seen":"2026-10-24T08:54:58Z"},{"id":"anthropics/claude-code#94718","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":94718,"title":"MCP client rejects tool calls omitting optional parameters; deadlocks on tools with mutually-exclusive parameter groups","url":"https://github.com/anthropics/claude-code/issues/94718","state":"closed","state_reason":"duplicate","labels":["bug","has repro","platform:macos","area:mcp"],"created":"2026-09-16T09:34:14Z","updated":"2026-10-11T23:39:59Z","closed":"2026-10-11T23:39:59Z","reactions":7,"thumbs_up":7,"comments":8,"severity":"high","severity_reasons":["title:breakage"],"topic":"crash/hang","classes":["crash/hang","mcp/tools","self-host/headless","install/update"],"cosmetic_ui":false,"summary":"Every MCP tool call that omits an optional parameter is rejected with a Zod-style invalidtype / expected: \"nonoptional\", received: \"undefined\" error for that parameter — even though the tool's own JSON Schema declares a default for it…","versions":["8.4.2","8.4.3","2026.9.2","2026.6.4"],"error_snippets":["\"error\": \"MCP error -32602: Input validation error: Invalid arguments for tool ha_get_state: [","{\\\"code\\\":\\\"invalid_type\\\",\\\"expected\\\":\\\"nonoptional\\\",\\\"path\\\":[\\\"fields\\\"],\\\"message\\\":\\\"Invalid input: expected nonoptional, received undefined\\\"},","{\\\"code\\\":\\\"invalid_type\\\",\\\"expected\\\":\\\"nonoptional\\\",\\\"path\\\":[\\\"attribute_keys\\\"],\\\"message\\\":\\\"Invalid input: expected nonoptional, received undefined\\\"}","\"error\": \"These parameters apply only to service calls and must be omitted when ws_command is set: entity_id, result_fields, result_attribute_keys.\"","\"expected\": \"nonoptional\","],"workaround":{"text":"explicitly pass `fields` and `attribute_keys` (e.g. as a real list, or as the literal string `null` for list-typed fields — note this is not JSON `null`, it becomes the one-element list `[\"null\"]` server-side, which is itself a separate minor bug but harmless for read calls).","source":"issue body","url":"https://github.com/anthropics/claude-code/issues/94718"},"fixed_in":null,"score":65.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"NousResearch/hermes-agent#135344","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":135344,"title":"kanban dispatcher: _pid_recycled compares the composed worker fingerprint by exact equality, so live macOS workers read as recycled and get duplicated","url":"https://github.com/NousResearch/hermes-agent/issues/135344","state":"open","state_reason":null,"labels":["type/bug","comp/cli","comp/cron","P2","sweeper:risk-automation","os/macos"],"created":"2026-10-08T22:48:59Z","updated":"2026-10-20T00:51:37Z","closed":null,"reactions":1,"thumbs_up":1,"comments":6,"severity":"medium","severity_reasons":["class:always-on/daemon"],"topic":"always-on/daemon","classes":["always-on/daemon","crash/hang"],"cosmetic_ui":false,"summary":"hermescli/kanbandbdispatch.py::pidrecycled has a string branch for the composed worker fingerprint (\" | \") that compares the recorded value to a freshly computed one with EXACT equality. On hosts without /proc (macOS) the start-time half…","versions":[],"error_snippets":["real = _process_fingerprint(pid) # \"|179149919058\" on macOS: empty epoch, start time","except (TypeError, ValueError):","\"<instantiation-epoch>|<start-time>\"","`gateway/status.py` already documents this and already tolerates it: the integer branch of the same function uses `start_time_fingerprints_match` with a 200cs tolerance, and its comment names the macOS `kern.boottime` ad","Keep the epoch half exact, because it is the reboot witness, and hand the start-time halves to the same tolerance comparator the integer branch already uses:"],"workaround":null,"fixed_in":null,"score":65.2,"first_seen":"2026-10-18T04:32:26Z"},{"id":"kirodotdev/Kiro#11677","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":11677,"title":"kiro-cli MCP OAuth fails with \"Authorization server response missing required issuer\": the loopback callback handler drops the RFC 9207 iss parameter","url":"https://github.com/kirodotdev/Kiro/issues/11677","state":"closed","state_reason":"completed","labels":["os: mac","mcp","auth","pending-maintainer-response"],"created":"2026-09-29T18:37:31Z","updated":"2026-10-21T18:10:50Z","closed":"2026-10-21T18:10:50Z","reactions":2,"thumbs_up":2,"comments":10,"severity":"high","severity_reasons":["title:breakage"],"topic":"self-host/headless","classes":["self-host/headless","auth/session","mcp/tools","crash/hang"],"cosmetic_ui":false,"summary":"kiro-cli 2.24.0 on macOS (Apple Silicon) Remote MCP server over streamable HTTP, OAuth 2.1 with dynamic client registration and PKCE (S256) The authorization server's metadata has \"authorizationresponseissparametersupported\": true The same…","versions":["2.24.0","2.1","2.1.284","0.158.0"],"error_snippets":["rmcp::transport::auth: start exchange code for token: \"<redacted>\"","agent::agent: forced MCP auth failed; reloading under normal flow target=demo","error=Authorization server response missing required issuer: expected https://<server>","The browser shows \"You can close this page now\". kiro-cli exchanges the code at the token endpoint and the server status becomes `running`.","The browser shows \"You can close this page now\". kiro-cli never calls the token endpoint. The server status stays `auth-required`, and kiro-cli starts a new registration and a new flow."],"workaround":null,"fixed_in":null,"score":65.1,"first_seen":"2026-10-19T18:38:33Z"},{"id":"kirodotdev/Kiro#12114","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":12114,"title":"Workflow run stuck \"running\" on a stale pr-loop watch — cannot be aborted (update_workflow fails to load saved state), survives window reload and multiple restarts","url":"https://github.com/kirodotdev/Kiro/issues/12114","state":"open","state_reason":null,"labels":["pending-triage","os: mac","hooks","theme:unexpected-error"],"created":"2026-10-25T18:51:12Z","updated":"2026-10-25T18:52:44Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"medium","severity_reasons":["class:install/update","class:crash/hang"],"topic":"install/update","classes":["install/update","crash/hang"],"cosmetic_ui":false,"summary":"Component: Workflow engine / orchestration (background workflow runs) Severity: Low functional impact (run is inert), but it's an unkillable orphaned run that clutters state and can't be cleaned up through any available in-product…","versions":["1.4.26"],"error_snippets":["Component: Workflow engine / orchestration (background workflow runs)","Severity: Low functional impact (run is inert), but it's an unkillable orphaned run that clutters state and can't be cleaned up through any available in-product mechanism.","Expected behavior A github-pr watch should reach a terminal condition when the watched PR is merged/closed (branch deleted), allowing the run to complete; and failing that, update_workflow { status: \"aborted\" } should fo","update_workflow abort fails every time with: \"Stop could not be applied: the workflow's saved state exists but could not be loaded into this process. Inspect the workflow before retrying.\"","The error's own remedy (\"inspect before retrying\") doesn't help — inspect_workflow succeeds (it can read the saved state) but the subsequent abort still can't load that state into the process to mutate it, suggesting the"],"workaround":null,"fixed_in":null,"score":64.8,"first_seen":"2026-10-26T10:12:02Z"},{"id":"omacom/omarchy#12187","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":12187,"title":"Update to 4.0.4: NVIDIA hybrid laptop hard-freezes ~5 s after login once linux-omarchy is the default kernel (prebuilt nvidia-open has no modules for it)","url":"https://github.com/omacom/omarchy/issues/12187","state":"open","state_reason":null,"labels":["bug","compatibility"],"created":"2026-09-16T21:13:48Z","updated":"2026-10-12T18:09:10Z","closed":null,"reactions":1,"thumbs_up":1,"comments":4,"severity":"high","severity_reasons":["title:breakage"],"topic":"install/update","classes":["install/update","crash/hang","auth/session","always-on/daemon"],"cosmetic_ui":false,"summary":"Omarchy 4.0.3-1 → 4.0.4-1 (update ran 2026-09-16; 4.0.4 released 2026-09-15) - Lenovo laptop, hybrid graphics: Intel iGPU + NVIDIA RTX 4070 Max-Q (AD106M, 10DE:2860), supergfxctl/supergfxd + nvidia-powerd active, Hyprland/Wayland - Driver…","versions":["4.0.4","4.0.3-1","4.0.4-1","610.57.04-14"],"error_snippets":["modules exist only under /usr/lib/modules/7.2.3-arch1-3/extramodules/ (built for the stock kernel); there are no NVIDIA modules for","Boot correlation (all frozen boots were on 7.2.5-3-omarchy, all working boots on 7.2.3-arch1-3; journals end abruptly with zero kernel errors):","Identical lines at login in every frozen boot:","systemd-modules-load: Failed to find module 'nvidia_uvm'","supergfxd: ERROR supergfxctl::controller Action thread errored: The module nvidia_drm is missing"],"workaround":null,"fixed_in":null,"score":64.5,"first_seen":"2026-10-24T08:54:58Z"},{"id":"google-gemini/gemini-cli#28717","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":28717,"title":"[Bug]: Google AI Pro sign-in silently falls back to API key onboarding in v0.54.0","url":"https://github.com/google-gemini/gemini-cli/issues/28717","state":"open","state_reason":null,"labels":["priority/p2","area/security","status/bot-triaged","kind/bug"],"created":"2026-08-07T00:22:18Z","updated":"2026-09-24T15:40:29Z","closed":null,"reactions":5,"thumbs_up":5,"comments":3,"severity":"critical","severity_reasons":["label:security"],"topic":"auth/session","classes":["auth/session","install/update"],"cosmetic_ui":false,"summary":"On Gemini CLI v0.54.0, selecting \"Sign in with Google\" and completing the browser flow with a personal Google account that has an active Google AI Pro subscription returns the CLI to the \"Enter Gemini API Key\" screen. The CLI silently…","versions":["0.54.0"],"error_snippets":["Enter Gemini API Key","On Gemini CLI v0.54.0, selecting \"Sign in with Google\" and completing the browser flow with a personal Google account that has an active Google AI Pro subscription returns the CLI to the \"Enter Gemini API Key\" screen.","The CLI provides no deprecation or migration explanation. Instead, it silently redirects the user to an API-key flow, which has a separate quota and potentially separate usage-based billing.","2. Ensure none of these variables are present:","3. Ensure `/etc/gemini-cli/system-defaults.json` does not exist."],"workaround":null,"fixed_in":null,"score":64.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"anomalyco/opencode#49847","harness":"opencode","harness_name":"OpenCode","repo":"anomalyco/opencode","number":49847,"title":"openai: ChatGPT OAuth requests use Zen API key","url":"https://github.com/anomalyco/opencode/issues/49847","state":"open","state_reason":null,"labels":[],"created":"2026-09-18T19:37:58Z","updated":"2026-10-11T22:38:48Z","closed":null,"reactions":3,"thumbs_up":3,"comments":9,"severity":"medium","severity_reasons":["class:auth/session"],"topic":"auth/session","classes":["auth/session"],"cosmetic_ui":false,"summary":"incorrectly binds the OpenAI provider to the OpenCode Zen integration, causing a Zen API key to be sent to ChatGPT’s OAuth-only Codex endpoint. OpenAI rejects the key with “API keys are not supported by this endpoint.” The correct OpenAI…","versions":["2.0.8"],"error_snippets":["incorrectly binds the OpenAI provider to the OpenCode Zen integration, causing a Zen API key to be sent to ChatGPT’s OAuth-only Codex endpoint. OpenAI rejects the key with “API keys are not supported by this endpoint.” T","> API keys are not supported by this endpoint.","The request goes to the correct ChatGPT Codex endpoint and includes the correct chatgpt-account-id, but its Authorization header contains the OpenCode Zen API key instead of the OpenAI OAuth token.","Expected: requests routed to the ChatGPT Codex endpoint use the credential from the openai integration.","API keys are not supported by this endpoint."],"workaround":null,"fixed_in":null,"score":64.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openclaw/openclaw#155191","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":155191,"title":"Native memory leak in 2026.9.5 — RSS grows ~1 GiB per 30 s while V8 heap stays stable","url":"https://github.com/openclaw/openclaw/issues/155191","state":"open","state_reason":null,"labels":["clawsweeper:no-new-fix-pr","clawsweeper:needs-maintainer-review","clawsweeper:needs-info","impact:crash-loop","P0","issue-rating: 🦐 gold shrimp"],"created":"2026-09-21T20:57:37Z","updated":"2026-10-11T23:18:59Z","closed":null,"reactions":2,"thumbs_up":1,"comments":8,"severity":"critical","severity_reasons":["label:p0"],"topic":"crash/hang","classes":["crash/hang","always-on/daemon","install/update","network"],"cosmetic_ui":false,"summary":"On OpenClaw 2026.9.5 (build ec9c1a1, installed via npm on 2026-09-18), the gateway node process exhibits a sawtooth RSS pattern: process RSS jumps by 1–2 GiB within 30-second windows while V8 heapUsed stays between 786 MB and 1 GB. The…","versions":["2026.9.5","2026.9.4","26.8.2","2.25"],"error_snippets":["rssBytes: 425 → 2 090 MB (5× swing)","rssBefore: 2 521 432 064","rssAfter: 3 081 453 568 ← RSS GREW 560 MB during sampling","selfSize 66 464 B per call","13:06:52 (node:44583) MaxListenersExceededWarning: 11 exit listeners added to [process]. MaxListeners is 10."],"workaround":null,"fixed_in":null,"score":63.9,"first_seen":"2026-10-18T04:33:05Z"},{"id":"openclaw/openclaw#110186","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":110186,"title":"v2026.7.2 STRICT migration fails on databases with BLOB-stored TEXT values and PRIMARY KEY constraints","url":"https://github.com/openclaw/openclaw/issues/110186","state":"open","state_reason":null,"labels":["clawsweeper:no-new-fix-pr","clawsweeper:source-repro","clawsweeper:linked-pr-open","impact:crash-loop","P0","issue-rating: 🦞 diamond lobster","impact:ux-release-blocker"],"created":"2026-07-17T21:32:03Z","updated":"2026-10-15T15:04:55Z","closed":null,"reactions":4,"thumbs_up":3,"comments":1,"severity":"critical","severity_reasons":["label:p0"],"topic":"install/update","classes":["install/update","crash/hang","always-on/daemon"],"cosmetic_ui":false,"summary":"After upgrading to OpenClaw 2026.7.2-beta.2, openclaw gateway restart (or openclaw doctor --fix) fails with: followed by a chain of noncanonical X schema that cannot be repaired errors. Gateway cannot start, systemd enters a failed state…","versions":["2026.7.2","2026.7.2-beta.2","2026.7.1"],"error_snippets":["Failed migrating shared state database schema at /home/foxrain/.openclaw/state/openclaw.sqlite:","Error: SQLite table X changes rowid storage from Y to Z; refusing an identity-changing STRICT migration","config_path TEXT NOT NULL PRIMARY KEY, -- composite / non-integer PK","noncanonical X schema that cannot be repaired","followed by a chain of `noncanonical X schema that cannot be repaired` errors. Gateway cannot start, systemd enters a `failed` state, port 19999 stays free."],"workaround":null,"fixed_in":null,"score":63.6,"first_seen":"2026-10-18T04:30:03Z"},{"id":"NousResearch/hermes-agent#122299","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":122299,"title":"kanban dispatcher: worker spawn argv guard checks importability in parent, unsound for bare-interpreter child (ModuleNotFoundError: hermes_cli)","url":"https://github.com/NousResearch/hermes-agent/issues/122299","state":"closed","state_reason":"completed","labels":["type/bug","comp/cli","comp/cron","P2","sweeper:risk-compatibility","area/install-update","area/profiles"],"created":"2026-09-25T04:50:46Z","updated":"2026-09-28T10:37:12Z","closed":"2026-09-28T10:37:12Z","reactions":5,"thumbs_up":5,"comments":13,"severity":"medium","severity_reasons":["engagement 18","class:always-on/daemon","class:install/update"],"topic":"always-on/daemon","classes":["always-on/daemon","install/update","crash/hang"],"cosmetic_ui":false,"summary":"resolvehermesargv() in hermescli/kanbandbdispatch.py decides the kanban worker's spawn argv using an importability check evaluated in the gateway (parent) process, but that result does not hold for the freshly-spawned child. On installs…","versions":[],"error_snippets":["cd /tmp && ~/.hermes/tools/python-3.14.7+.../bin/python3 -m hermes_cli.main --version; echo exit=$?","# -> ModuleNotFoundError: No module named 'hermes_cli'; exit=1","cd /tmp && ~/.hermes/hermes-agent/.hermes/bin/hermes --version; echo exit=$?","sys.executable","package, every worker crashes on startup with"],"workaround":{"text":"Pin `HERMES_BIN` to the shim for the gateway via a systemd drop-in (survives self-update, unlike a source patch):","source":"issue body","url":"https://github.com/NousResearch/hermes-agent/issues/122299"},"fixed_in":null,"score":63.4,"first_seen":"2026-10-18T04:29:32Z"},{"id":"NousResearch/hermes-agent#135500","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":135500,"title":"[Bug]: Intermittent server_error with openai-codex across models on Windows Desktop","url":"https://github.com/NousResearch/hermes-agent/issues/135500","state":"open","state_reason":null,"labels":["type/bug","comp/agent","platform/discord","provider/openai","P2","needs-repro","comp/desktop","os/windows"],"created":"2026-10-09T08:08:09Z","updated":"2026-10-17T07:47:32Z","closed":null,"reactions":2,"thumbs_up":2,"comments":3,"severity":"medium","severity_reasons":["class:self-host/headless"],"topic":"self-host/headless","classes":["self-host/headless","always-on/daemon","auth/session","mcp/tools"],"cosmetic_ui":false,"summary":"Hermes Agent intermittently fails with APIError and servererror when using the openai-codex provider with gpt-6.1-sol. The issue occurs in both Hermes Desktop and Discord Gateway on Windows 11. Some requests complete successfully…","versions":[],"error_snippets":["Provider unavailable (server_error)","I can provide sanitized diagnostic output and relevant log excerpts upon request.","Relevant sanitized log excerpt:","status=error","error_retained=True"],"workaround":null,"fixed_in":null,"score":63.2,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aaif-goose/goose#13011","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":13011,"title":"Desktop (Windows, 1.53.0): turn-final assistant messages ([thinking + text], no tool call) missing from chat view, while fully stored and re-played by the server","url":"https://github.com/aaif-goose/goose/issues/13011","state":"open","state_reason":null,"labels":[],"created":"2026-10-19T23:40:55Z","updated":"2026-10-23T00:30:40Z","closed":null,"reactions":0,"thumbs_up":0,"comments":2,"severity":"medium","severity_reasons":["class:self-host/headless","class:mcp/tools"],"topic":"self-host/headless","classes":["self-host/headless","mcp/tools","auth/session"],"cosmetic_ui":false,"summary":"Desktop (Windows, 1.53.0): turn-final assistant messages ([thinking + text], no tool call) missing from chat view, while fully stored and re-played by the server Goose Desktop 1.53.0, Windows 11 Pro; provider ollamacloud Models…","versions":["1.53.0","5.3-flash","1.51"],"error_snippets":["(551 message rows),","In tool-loop turns the chat view shows the expanded thinking block, tool requests and tool output — then the turn visually ends. The turn-final assistant message (content `[thinking + text]`, **no trailing toolRequest**)","Observed correlation (small sample — observed, not established): **both** invisible messages are `[thinking + text]` with **no** trailing toolRequest, while every intermediate message that did render carried a toolReques","1. **Model/provider never failed.** `usage_ledger` for `20261019_5`: 235 requests, 0 with `output_tokens < 40`; no error entries for LLM traffic in serve logs.","2. **The message streamed fully (18002).** `llm_request.N.jsonl` shows that request as 867 streamed deltas ending in a finalize usage record (output 2,093 tokens). The 22:59 UTC 17901 roll was already overwritten before "],"workaround":null,"fixed_in":null,"score":63.2,"first_seen":"2026-10-20T09:28:14Z"},{"id":"NousResearch/hermes-agent#135965","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":135965,"title":"[Bug]: Telegram DM topic mode: typing in All to start a new topic hijacks the actively running topic","url":"https://github.com/NousResearch/hermes-agent/issues/135965","state":"closed","state_reason":"completed","labels":["type/bug","comp/gateway","platform/telegram","P3","sweeper:risk-session-state","sweeper:risk-message-delivery"],"created":"2026-10-10T04:20:10Z","updated":"2026-10-20T04:48:48Z","closed":"2026-10-20T04:47:56Z","reactions":12,"thumbs_up":12,"comments":5,"severity":"medium","severity_reasons":["engagement 17","class:always-on/daemon"],"topic":"always-on/daemon","classes":["always-on/daemon"],"cosmetic_ui":false,"summary":"In Telegram DM topic mode, when a topic is actively running, typing a new message in the All lane to start a new topic: creates a new (empty, dead) topic, and the message text lands in the actively running topic/session instead — the new…","versions":[],"error_snippets":["1. creates a new (empty, dead) topic, and","2. the message text lands in the **actively running** topic/session instead — the new topic stays empty, and the running conversation receives a message it wasn't meant to have.","1. Enable Telegram DM topic mode (e.g. `/topic` in the DM) and have at least one running/active topic.","3. Telegram creates the empty topic; the message you typed appears in the **previously active** topic instead.","The message is delivered to the lobby (no `message_thread_id`, or General id `\"1\"`), and `_recover_telegram_topic_thread_id` pins **any** lobby-shaped message to the user's most-recently-bound topic — so the new topic st"],"workaround":null,"fixed_in":null,"score":63.0,"first_seen":"2026-10-18T04:29:32Z"},{"id":"openclaw/openclaw#159638","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":159638,"title":"Gateway boots ~7 short-lived SQLite store/read workers per minute (512 MB each), keeping V8 threads at ~80% CPU","url":"https://github.com/openclaw/openclaw/issues/159638","state":"open","state_reason":null,"labels":["P1","clawsweeper:no-new-fix-pr","clawsweeper:needs-maintainer-review","issue-rating: 🦪 silver shellfish","impact:other"],"created":"2026-09-27T11:59:35Z","updated":"2026-10-18T01:35:56Z","closed":null,"reactions":2,"thumbs_up":1,"comments":7,"severity":"high","severity_reasons":["label:p1"],"topic":"always-on/daemon","classes":["always-on/daemon","install/update","mcp/tools"],"cosmetic_ui":false,"summary":"On an idle-ish 2026.9.6 gateway, SQLite store/read workers are created and torn down about 7 times per minute. Each one boots the full worker runtime (about 2,000 module files, all plugin manifests) with a 512 MB heap limit and exits after…","versions":["2026.9.6"],"error_snippets":["13:23:36 START state-read 13:23:41 EXIT","13:23:36 START sqlite-store 13:23:41 EXIT","13:23:51 START sqlite-store 13:23:53 EXIT","13:24:07 START sqlite-store 13:24:11 EXIT","13:24:13 START session-transcript 13:24:38 EXIT"],"workaround":null,"fixed_in":null,"score":63.0,"first_seen":"2026-10-18T04:33:05Z"},{"id":"anthropics/claude-code#101865","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":101865,"title":"Claude Code webview blank in WSL after VS Code update, 401 on webview resources","url":"https://github.com/anthropics/claude-code/issues/101865","state":"closed","state_reason":"completed","labels":["bug","platform:vscode","platform:wsl"],"created":"2026-10-13T19:17:07Z","updated":"2026-10-16T07:30:07Z","closed":"2026-10-16T07:30:07Z","reactions":5,"thumbs_up":5,"comments":2,"severity":"medium","severity_reasons":["class:self-host/headless","class:install/update"],"topic":"self-host/headless","classes":["self-host/headless","install/update","auth/session"],"cosmetic_ui":false,"summary":"After updating VS Code, the Claude Code panel is blank when connected to WSL (Ubuntu). The extension activates normally and authentication works (Output log shows no errors), but the webview fails to load its files: GET…","versions":["2.1.296-linux","2.1.296","1.141.1","1.0.92-4.unstable.r37397886721.gad270aa"],"error_snippets":["nes-ex-cf-euporie-03:31542868","codex-agent-host-stable-editor:31596219","After updating VS Code, the Claude Code panel is blank when connected to WSL (Ubuntu). The extension activates normally and authentication works (Output log shows no errors), but the webview fails to load its files:","GET https://vscode-remote+wsl-002bubuntu.vscode-resource.vscode-cdn.net/home/ /.vscode-server/extensions/anthropic.claude-code-2.1.296-linux-x64/webview/index.js net::ERR_ABORTED 401 (Unauthorized)","GET .../webview/index.css net::ERR_ABORTED 401 (Unauthorized)"],"workaround":null,"fixed_in":null,"score":63.0,"first_seen":"2026-10-18T04:33:44Z"},{"id":"github/copilot-cli#5236","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":5236,"title":"Windows: Entra broker token cache stays 0 bytes after MCP sign-in, so every new process re-prompts","url":"https://github.com/github/copilot-cli/issues/5236","state":"open","state_reason":null,"labels":["area:platform-windows","area:mcp"],"created":"2026-10-22T18:48:08Z","updated":"2026-10-25T18:57:55Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"medium","severity_reasons":["class:auth/session","class:mcp/tools"],"topic":"auth/session","classes":["auth/session","mcp/tools","self-host/headless"],"cosmetic_ui":false,"summary":"After I sign in through the Microsoft Entra account broker for an HTTP MCP server, the token cache file stays empty. Every new process shows the account picker again. A /restart in the same process does not prompt…","versions":["1.0.103"],"error_snippets":["~\\.copilot\\entra-broker-accounts\\<id>.bin","~\\.copilot\\entra-broker-cache\\<id>.bin","{\"mcpServers\": {\"EngHub\": {\"type\": \"http\", \"url\": \"https://mcp.eng.ms\", \"tools\": [\"*\"]}}}","Authenticating: EngHub","~\\.copilot\\entra-broker-cache\\"],"workaround":null,"fixed_in":null,"score":63.0,"first_seen":"2026-10-23T09:23:24Z"},{"id":"openclaw/openclaw#167477","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":167477,"title":"Setup inventory stays retired after a config hot reload: setup-registry, setup-inference, model runtime refresh, subagent announce, and chat.metadata fail until restart","url":"https://github.com/openclaw/openclaw/issues/167477","state":"open","state_reason":null,"labels":["P1","clawsweeper:no-new-fix-pr","clawsweeper:needs-maintainer-review","impact:message-loss","impact:auth-provider","issue-rating: 🦪 silver shellfish"],"created":"2026-10-08T22:07:45Z","updated":"2026-10-22T08:24:36Z","closed":null,"reactions":2,"thumbs_up":1,"comments":2,"severity":"high","severity_reasons":["label:p1"],"topic":"auth/session","classes":["auth/session","always-on/daemon","install/update","crash/hang","mcp/tools"],"cosmetic_ui":false,"summary":"After a config hot reload of an agents.entries. change, the plugin setup inventory was retired but several callers kept using it. For about 1.5 hours, until a Gateway restart, they failed with Plugin setup inventory has retired / Plugin…","versions":["1.5","2026.9.6"],"error_snippets":["13:30:24 [reload] config change detected; evaluating reload (agents.entries.code-orchestrator)","13:30:57 [secrets] [SECRETS_DEGRADED] cold provider:deepseek: secret reference was not found.","13:31:02 [reload] config hot reload applied (agents.entries.code-orchestrator)","13:35:04 [plugins/setup-registry] plugin setup [browser] setup-entry-load-failed: setup entry failed to load from …/dist/extensions/browser/setup-api.js: Plugin browser setup inventory has retired","(same for [canvas], [elevenlabs], [xai]; repeats every few minutes, 11 rounds through 14:13)"],"workaround":null,"fixed_in":null,"score":62.8,"first_seen":"2026-10-25T09:11:37Z"},{"id":"kirodotdev/Kiro#12035","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":12035,"title":"kiro-cli v3: local command sandbox blocks the user-database lookup on macOS, so ssh and ssh-keygen can't run","url":"https://github.com/kirodotdev/Kiro/issues/12035","state":"open","state_reason":null,"labels":["ssh","trusted-commands","pending-maintainer-response","cli","bug"],"created":"2026-10-20T08:40:24Z","updated":"2026-10-22T16:08:50Z","closed":null,"reactions":0,"thumbs_up":0,"comments":3,"severity":"medium","severity_reasons":["class:self-host/headless"],"topic":"self-host/headless","classes":["self-host/headless","network"],"cosmetic_ui":false,"summary":"macOS 27.0.1 (arm64) kiro-cli 2.32.0 With the local command sandbox enabled, ssh and ssh-keygen exit immediately with No user exists for uid 501. OpenSSH calls getpwuid() on startup and gives up when it fails, so the Seatbelt profile…","versions":["2.32.0"],"error_snippets":["With the local command sandbox enabled, ssh and ssh-keygen exit immediately with No user exists for uid 501. OpenSSH calls getpwuid() on startup and gives up when it fails, so the Seatbelt profile appears to block the di","1. In an empty project directory, create .kiro/settings/sandbox.json:","{ \"enabled\": true, \"networkAccess\": \"COMMON_DEPENDENCIES\" }","2. Start a V3 session there: kiro-cli chat --agent-engine v3. Kiro reports Sandbox is on. Current network mode: common dependencies.","3. Ask the agent to run ssh -F /dev/null -o BatchMode=yes -T git@github.com; echo $?. It prints No user exists for uid 501 and exits with 255."],"workaround":null,"fixed_in":null,"score":62.8,"first_seen":"2026-10-20T09:28:14Z"},{"id":"google-gemini/gemini-cli#23362","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":23362,"title":"Bug Report: 429 Resource Exhausted (MODEL_CAPACITY_EXHAUSTED) in Gemini CLI","url":"https://github.com/google-gemini/gemini-cli/issues/23362","state":"closed","state_reason":"not_planned","labels":["priority/p2","area/core","Stale","status/possible-duplicate","status/need-retesting","status/bot-triaged","kind/bug","effort/small"],"created":"2026-03-21T08:54:05Z","updated":"2026-09-16T10:14:27Z","closed":"2026-08-05T02:20:34Z","reactions":6,"thumbs_up":6,"comments":18,"severity":"medium","severity_reasons":["engagement 24","class:network"],"topic":"network","classes":["network","self-host/headless","auth/session"],"cosmetic_ui":false,"summary":"I am use Gemini Pro account login, but keep met the following \"Resource Exhausted\" issue. if i use API key login, then dont have this issue. How to check if my account be flagged abuse? no 429 isse. Client Information Run gemini to enter…","versions":["0.34.0"],"error_snippets":["\"code\": 429,","\"@type\": \"type.googleapis.com/google.rpc.ErrorInfo\",","to enter the interactive CLI, then run the","I am use Gemini Pro account login, but keep met the following \"Resource Exhausted\" issue. if i use API key login, then dont have this issue.","no 429 isse."],"workaround":null,"fixed_in":null,"score":62.4,"first_seen":"2026-10-19T18:38:33Z"},{"id":"kirodotdev/Kiro#11981","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":11981,"title":"Task-execution tools (task_get/task_update/task_status) and spec-task-execution dispatch intermittently drop out mid-session","url":"https://github.com/kirodotdev/Kiro/issues/11981","state":"open","state_reason":null,"labels":["pending-maintainer-response"],"created":"2026-10-16T19:22:58Z","updated":"2026-10-21T10:08:52Z","closed":null,"reactions":0,"thumbs_up":0,"comments":5,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"What happens: During a spec task-execution session, the task-status tools and the spec-task-execution subagent dispatch repeatedly become unavailable (\"Tool X is not available\"), then come back after a restart, then drop again later in the…","versions":["1.3.11"],"error_snippets":["What happens: During a spec task-execution session, the task-status tools and the spec-task-execution subagent dispatch repeatedly become unavailable (\"Tool X is not available\"), then come back after a restart, then drop","Can't move tasks through queued → in_progress → completed, so tasks.md checkboxes drift out of sync with the task-status service.","spec-task-execution dispatch is gated on a task being marked in_progress via the task tooling — when that tooling is down, the gate rejects every dispatch (\"mark this exact task as in_progress and wait… then retry\"), whi","Burns tokens: each dropout triggers retries, manual workarounds (editing checkboxes by hand), and re-verification.","Later in the same session they dropped again, and the dispatch gate blocked all spec-task-execution work."],"workaround":null,"fixed_in":null,"score":62.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aaif-goose/goose#13022","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":13022,"title":"Add harness-level progressive MCP tool discovery","url":"https://github.com/aaif-goose/goose/issues/13022","state":"open","state_reason":null,"labels":[],"created":"2026-10-20T13:52:05Z","updated":"2026-10-21T00:48:32Z","closed":null,"reactions":0,"thumbs_up":0,"comments":3,"severity":"medium","severity_reasons":["class:mcp/tools"],"topic":"mcp/tools","classes":["mcp/tools","self-host/headless","network"],"cosmetic_ui":false,"summary":"Add harness-level progressive tool discovery so Goose can keep most MCP tool schemas out of the model context until they are relevant. A small always-visible discovery tool could search across configured MCP servers using server/tool…","versions":[],"error_snippets":["A small always-visible discovery tool could search across configured MCP servers using server/tool hints, activate matching tools for the session, and refresh the model-visible catalog. MCP servers that support dynamic c","Large MCPs (GitHub, Google Drive, etc.) can create tool-schema explosion. Buzz has the same issue: Buzz PR #1329 marks tools as always-loaded or search-loaded for harnesses that support deferred tools, but Goose currentl","Related: #10955 (progressive extension activation for external harnesses)."],"workaround":null,"fixed_in":null,"score":61.9,"first_seen":"2026-10-26T10:12:02Z"},{"id":"google-gemini/gemini-cli#29930","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":29930,"title":"ACP: run_shell_command output is dropped (empty tool_call_update content) when the shell returns AnsiOutput","url":"https://github.com/google-gemini/gemini-cli/issues/29930","state":"open","state_reason":null,"labels":["area/non-interactive","status/bot-triaged","effort/small"],"created":"2026-10-25T12:29:04Z","updated":"2026-10-25T13:12:58Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update","auth/session"],"cosmetic_ui":false,"summary":"In ACP mode, a runshellcommand call that prints output ends with a toolcallupdate whose content is empty, so an ACP client never sees the command's output. toToolCallContent in packages/cli/src/acp/acpUtils.ts forwards returnDisplay when…","versions":["0.65.0"],"error_snippets":["{\"sessionUpdate\":\"tool_call\",\"status\":\"in_progress\",\"title\":\"echo hello-acp\",\"content\":[{\"type\":\"content\",\"content\":{\"type\":\"text\",\"text\":\"[current working directory …] (repro)\"}}]}","{\"sessionUpdate\":\"tool_call_update\",\"status\":\"completed\",\"title\":\"echo hello-acp\",\"content\":[]}",") whenever that is non-empty, and to","invocation.execute","{\"command\":\"echo hello-acp\"}"],"workaround":null,"fixed_in":null,"score":61.8,"first_seen":"2026-10-26T10:12:02Z"},{"id":"google-gemini/gemini-cli#29936","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":29936,"title":"test(cli): remove redundant lowercase restart test in AdminSettingsChangedDialog","url":"https://github.com/google-gemini/gemini-cli/issues/29936","state":"open","state_reason":null,"labels":["status/need-triage","area/platform"],"created":"2026-10-25T19:21:12Z","updated":"2026-10-25T21:44:13Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"medium","severity_reasons":["class:crash/hang"],"topic":"crash/hang","classes":["crash/hang","network"],"cosmetic_ui":false,"summary":"AdminSettingsChangedDialog.test.tsx tests the lowercase restart key twice: The standalone test at lines 26–41 renders the dialog, writes 'r' to stdin, and checks that handleRestartMock was called. The parameterized test at lines 43–58 uses…","versions":[],"error_snippets":["it.each(['r', 'R'])"],"workaround":null,"fixed_in":null,"score":61.8,"first_seen":"2026-10-26T10:12:02Z"},{"id":"omacom/omarchy#15924","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":15924,"title":"Blur behind GTK4 windows is gone after the gtk4 4.24 update (only Glassy keeps it)","url":"https://github.com/omacom/omarchy/issues/15924","state":"open","state_reason":null,"labels":[],"created":"2026-10-19T04:57:52Z","updated":"2026-10-23T05:28:57Z","closed":null,"reactions":1,"thumbs_up":1,"comments":1,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update","always-on/daemon"],"cosmetic_ui":false,"summary":"Intel Core i5-1335U / Iris Xe (HP Pavilion 15-eg3xxx), Omarchy 4.0.5, Hyprland 0.56.2, Wayland, gtk4 1:4.24.1 After the 2026-10-18 update, Hyprland stopped blurring the background behind Ghostty. The window is still translucent and the…","versions":["4.24","4.0.5","0.56.2","4.24.1"],"error_snippets":["return !surface->m_blurRegion.empty(); // empty region => no blur","zswap.enabled=0 rootflags=subvol=@ rw rootfstype=btrfs rtc_cmos.use_acpi_alarm=1","Topology cpus 1x dies 1 clusters 4 cores 10 threads 12 mt 2 tpc 2 st 8 smt enabled cache","Speed (MHz) avg 794 min/max 400/4600:3400 scaling driver intel_pstate governor powersave cores","Type spectre_v2 mitigation Enhanced / Automatic IBRS; IBPB: conditional; PBRSB-eIBRS: SW"],"workaround":null,"fixed_in":null,"score":61.8,"first_seen":"2026-10-25T09:11:37Z"},{"id":"openclaw/openclaw#85030","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":85030,"title":"[Bug]: MCP tools not injected into subagent (sessions_spawn) sessions — `bundle-mcp` + per-tool subagent allowlist + per-agent allowlist all ignored, subagent system prompt and tool schema receive only built-ins","url":"https://github.com/openclaw/openclaw/issues/85030","state":"closed","state_reason":"completed","labels":["bug","bug:behavior","P1","clawsweeper:no-new-fix-pr","clawsweeper:needs-maintainer-review","clawsweeper:needs-product-decision","clawsweeper:needs-security-review","clawsweeper:source-repro","impact:session-state","impact:security","issue-rating: 🦞 diamond lobster"],"created":"2026-05-21T16:45:41Z","updated":"2026-10-01T19:59:19Z","closed":"2026-09-13T13:46:52Z","reactions":7,"thumbs_up":6,"comments":16,"severity":"critical","severity_reasons":["label:security"],"topic":"mcp/tools","classes":["mcp/tools","always-on/daemon","self-host/headless","auth/session"],"cosmetic_ui":false,"summary":"Behavior bug (incorrect output/state without crash) MCP tool schemas registered via mcp.servers are not injected into sessions spawned by sessionsspawn, regardless of how the documented exposure mechanism is configured. The documented…","versions":[],"error_snippets":["- mcp.servers entry (Authorization redacted to \"Bearer ${MCP_KEY_OC}\")","- tools.subagents block (shows bundle-mcp + per-tool MCP entries present)","- per-agent tools blocks for delegator + 3 target agents (shows per-agent gst entries on targets)","`openclaw sessions export-trajectory`. Contains exactly 4 entries: exec, message, read, sessions_yield.","No MCP entries, no bundle-mcp, no gst-mcp__* names."],"workaround":null,"fixed_in":null,"score":61.7,"first_seen":"2026-10-19T18:38:33Z"},{"id":"github/copilot-cli#5222","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":5222,"title":"Startup reports disabled MCP servers as “blocked by policy” and repeats the warnings","url":"https://github.com/github/copilot-cli/issues/5222","state":"open","state_reason":null,"labels":["area:mcp"],"created":"2026-10-21T19:45:45Z","updated":"2026-10-22T05:25:56Z","closed":null,"reactions":1,"thumbs_up":1,"comments":0,"severity":"medium","severity_reasons":["class:self-host/headless","class:mcp/tools"],"topic":"self-host/headless","classes":["self-host/headless","mcp/tools"],"cosmetic_ui":false,"summary":"Copilot CLI displays organization-policy warnings at startup even though every configured third-party MCP server is explicitly disabled through disabledMcpServers in the user settings. The same pair of warnings appears twice during…","versions":["1.0.101"],"error_snippets":["Copilot CLI displays organization-policy warnings at startup even though every configured third-party MCP server is explicitly disabled through `disabledMcpServers` in the user settings. The same pair of warnings appears","Both messages appear twice in the observed startup. Every server named in the warning is already explicitly disabled in the user settings.","These observations explain the reporting behavior; **they do not establish that disabled servers are launched**. The exact trigger for the second initialization/reload has not been identified."],"workaround":null,"fixed_in":null,"score":61.4,"first_seen":"2026-10-26T10:12:02Z"},{"id":"google-gemini/gemini-cli#19936","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":19936,"title":"Stuck in loop of verification","url":"https://github.com/google-gemini/gemini-cli/issues/19936","state":"closed","state_reason":"completed","labels":["priority/p2","area/core","status/bot-triaged","kind/bug","effort/medium"],"created":"2026-02-22T19:15:19Z","updated":"2026-10-07T22:40:47Z","closed":"2026-10-07T22:40:47Z","reactions":5,"thumbs_up":5,"comments":19,"severity":"medium","severity_reasons":["engagement 24","class:crash/hang"],"topic":"crash/hang","classes":["crash/hang","auth/session","network"],"cosmetic_ui":false,"summary":"Asked to verify my pro account to continue, clicked link, verified, got taken to page that states: Authentication successful The authentication was successful, and the following products are now authorized to access your account: Gemini…","versions":["2.5"],"error_snippets":["to enter the interactive CLI, then run the","Then it tells me to hit enter once verified. I do that and it prompts me to go through verification again and again endlessley.","This is timed around losing access to 3, and only seeing 2.5 in models even after having used 3 since release.","I expected to verify once and continue using.","Run `gemini` to enter the interactive CLI, then run the `/about` command."],"workaround":null,"fixed_in":null,"score":61.3,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aws/amazon-q-developer-cli#3695","harness":"amazon-q-cli","harness_name":"Amazon Q Developer CLI","repo":"aws/amazon-q-developer-cli","number":3695,"title":"Social login (Google) has no profileArn; chat fails with \"profileArn is required but no profiles are available","url":"https://github.com/aws/amazon-q-developer-cli/issues/3695","state":"open","state_reason":null,"labels":[],"created":"2026-03-27T23:24:33Z","updated":"2026-10-22T21:34:00Z","closed":null,"reactions":1,"thumbs_up":1,"comments":7,"severity":"medium","severity_reasons":["class:auth/session"],"topic":"auth/session","classes":["auth/session","install/update"],"cosmetic_ui":false,"summary":"I have run q doctor in the affected terminal session I have run q restart and replicated the issue again macOS 26.3.1 (25D771280a) After logging in with Google (SocialGoogle), kiro-cli whoami should include a resolved Profile section…","versions":["1.28.2","48.00","5.9","3.5"],"error_snippets":["[environment]","Chat should work for social login users without requiring an enterprise profile configuration.","Email address","→ Shows SocialGoogle and email, but no profile","Observe error:"],"workaround":null,"fixed_in":null,"score":61.3,"first_seen":"2026-10-23T09:23:24Z"},{"id":"cline/cline#15147","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":15147,"title":"MCP: a remote server restart/deploy wedges the session — the client never re-initializes on a 404 `Session not found` (sdk/packages/core)","url":"https://github.com/cline/cline/issues/15147","state":"open","state_reason":null,"labels":[],"created":"2026-10-16T16:37:04Z","updated":"2026-10-17T06:58:07Z","closed":null,"reactions":0,"thumbs_up":0,"comments":3,"severity":"medium","severity_reasons":["class:self-host/headless","class:mcp/tools"],"topic":"self-host/headless","classes":["self-host/headless","mcp/tools","always-on/daemon","auth/session","network"],"cosmetic_ui":false,"summary":"A remote MCP server that speaks streamable HTTP (or SSE) is replaced under a live Cline session — a restart, a deploy/rollout, a session-TTL sweep. From that moment on, every tool call to that server fails, for the rest of the session…","versions":["2.0","1.30.0"],"error_snippets":["Streamable HTTP error: Error POSTing to endpoint: Not Found: Session not found","-H 'Accept: application/json, text/event-stream' \\","# HTTP/2 404","# Not Found: Session not found","attempt 1: threw \"Session not found\" session now deadbeef-dead-beef-dead-beefdeadbeef"],"workaround":{"text":"The only workaround is restarting Cline (in the CLI/TUI, restarting the hub daemon, which is what actually owns the connection).","source":"issue body","url":"https://github.com/cline/cline/issues/15147"},"fixed_in":null,"score":61.3,"first_seen":"2026-10-19T18:38:33Z"},{"id":"omacom/omarchy#14237","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":14237,"title":"Left DisplayPort monitor intermittently blacks out after wake on RTX 5080; 60 → 120 Hz reset workaround","url":"https://github.com/omacom/omarchy/issues/14237","state":"open","state_reason":null,"labels":["bug"],"created":"2026-10-04T13:01:58Z","updated":"2026-10-18T09:53:10Z","closed":null,"reactions":2,"thumbs_up":2,"comments":6,"severity":"medium","severity_reasons":["class:always-on/daemon"],"topic":"always-on/daemon","classes":["always-on/daemon"],"cosmetic_ui":false,"summary":"Omarchy 4.0.4-1; Hyprland 0.56.2-2 (Wayland) Kernel: 7.2.5-3-omarchy CPU: AMD Ryzen 9 7950X3D; motherboard: ASUS TUF GAMING X670E-PLUS WIFI GPU driving both monitors: NVIDIA GeForce RTX 5080 (MSI), nvidia-open-dkms / nvidia-utils…","versions":["4.0.4-1","0.56.2-2","610.57.04-1","59.951"],"error_snippets":["hyprctl eval 'hl.monitor({ output = \"DP-2\", mode = \"2560x1440@60\", position = \"0x0\", scale = 1, transform = 1 })'","hyprctl eval 'hl.monitor({ output = \"DP-2\", mode = \"2560x1440@120\", position = \"0x0\", scale = 1, transform = 1 })'","systemctl is-enabled","The left monitor intermittently goes black and comes back after sleep/wake, while the right monitor continues working normally. The desktop remains usable on the right display. Expected: both displays should recover and ","The sleep/wake association is based on my observations; the exact trigger has not yet been isolated between system suspend/resume and display idle/DPMS wake. The available current-boot journal does not record an actual s"],"workaround":null,"fixed_in":null,"score":61.3,"first_seen":"2026-10-19T18:38:33Z"},{"id":"kirodotdev/Kiro#11816","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":11816,"title":"/mcp auth for a remote OAuth server never surfaces the authorization URL (Google Workspace MCP, 403-not-401 servers)","url":"https://github.com/kirodotdev/Kiro/issues/11816","state":"open","state_reason":null,"labels":["mcp","auth"],"created":"2026-10-06T19:47:18Z","updated":"2026-10-07T17:30:24Z","closed":null,"reactions":0,"thumbs_up":0,"comments":4,"severity":"medium","severity_reasons":["class:self-host/headless","class:auth/session"],"topic":"self-host/headless","classes":["self-host/headless","auth/session","mcp/tools"],"cosmetic_ui":false,"summary":"macOS 15.7.9 (24G830) Kiro CLI version: 2.28.0 OS: macOS 15.7.9 (24G830) Server: Google Workspace first-party MCP (.mcp.googleapis.com, e.g. people.googleapis.com/mcp/v1), configured as a Desktop-app OAuth client with clientId +…","versions":["2.28.0"],"error_snippets":["HTTP 403: Method doesn't allow unregistered callers (callers without established identity). Please use API Key or other form of API consumer identity to call this API.","Running /mcp auth gpeople prints Forcing authentication for 'gpeople'… and then nothing further happens — no authorization URL is presented, no browser flow starts, no error. The OAuth URL (the _kiro.dev/mcp/oauth_reques","Expected /mcp auth should present the authorization URL (status panel, clipboard via Ctrl+Y, or an equivalent) so the browser consent flow can complete and a token can be minted.","Call any tool → observe the 403 \"unregistered callers\" error.","Impact Google Workspace MCP servers (Gmail/Drive/Docs/Sheets/Slides/Calendar/People) cannot be authenticated at all through this path. A correctly-configured, Developer-Preview-enrolled project with a valid Desktop OAuth"],"workaround":null,"fixed_in":null,"score":61.2,"first_seen":"2026-10-19T18:38:33Z"},{"id":"cline/cline#15031","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":15031,"title":"[JetBrains] cline-core 4.1.23 OOM after 6-12 h: suspended gRPC stream generator retains 1111 x 7.25 MB stateJson payloads","url":"https://github.com/cline/cline/issues/15031","state":"open","state_reason":null,"labels":["JetBrains","beta"],"created":"2026-10-12T14:44:51Z","updated":"2026-10-13T17:30:48Z","closed":null,"reactions":0,"thumbs_up":0,"comments":4,"severity":"high","severity_reasons":["title:breakage"],"topic":"install/update","classes":["install/update","crash/hang","mcp/tools"],"cosmetic_ui":false,"summary":"JetBrains Plugin 4.1.23 (JetBrains plugin core 2.0.7, extensionVariant: next) I am using a beta version of Cline cline-core running inside Android Studio grows until it hits the 8 GB heap limit and dies, killing the session mid-task. The…","versions":["4.1.23","7.25","2.0.7","8195.19"],"error_snippets":["Heap.20261012.213608.*.heapsnapshot 263,497,566 bytes","{\"version\":\"4.1.23\",\"extensionVariant\":\"next\",\"apiConfiguration\":{\"deepSeekApiKey\":...","<-.0 array (1111 elements)","<-.extension system / Context","93 Unhandled session event type: session_snapshot"],"workaround":null,"fixed_in":null,"score":61.1,"first_seen":"2026-10-19T18:38:33Z"},{"id":"kirodotdev/Kiro#12056","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":12056,"title":"Stuck on Working","url":"https://github.com/kirodotdev/Kiro/issues/12056","state":"open","state_reason":null,"labels":["pending-triage","pending-maintainer-response"],"created":"2026-10-21T18:34:31Z","updated":"2026-10-21T21:44:20Z","closed":null,"reactions":0,"thumbs_up":0,"comments":3,"severity":"medium","severity_reasons":["class:crash/hang"],"topic":"crash/hang","classes":["crash/hang"],"cosmetic_ui":false,"summary":"its working like about 10 min and its stuck then i was cancel the converstation and type continue and its stuck on working.... and i try this more then 10 time still stuck its working like about 10 min and its stuck then i was cancel the…","versions":["1.1.70"],"error_snippets":[],"workaround":null,"fixed_in":null,"score":60.6,"first_seen":"2026-10-22T09:27:06Z"},{"id":"kirodotdev/Kiro#11700","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":11700,"title":"Regression: `oauthScopes` still dropped from the OAuth authorization request on 2.26.0 (optional scope absent, tool calls 403) — reopens #8760 / #9123 after the CLI 2.12.0 scope-forwarding fix","url":"https://github.com/kirodotdev/Kiro/issues/11700","state":"open","state_reason":null,"labels":["os: mac","mcp","auth","cli"],"created":"2026-09-30T20:55:54Z","updated":"2026-10-01T19:56:18Z","closed":null,"reactions":0,"thumbs_up":0,"comments":4,"severity":"high","severity_reasons":["title:breakage"],"topic":"install/update","classes":["install/update","auth/session","mcp/tools","self-host/headless","crash/hang"],"cosmetic_ui":false,"summary":"macOS 26.7 (arm64, Apple Silicon) kiro-cli 2.26.0 — KAS 2.26.0, bundled node v22.22.2 Summary. A remote HTTP MCP server is configured with an explicit oauthScopes list that includes an optional provider scope. Kiro completes the OAuth…","versions":["2.26.0","2.12.0"],"error_snippets":["\"example-remote\": {","\"url\": \"https://mcp.example.com/mcp\",","\"clientId\": \"example-mcp-client\",","\"oauthScopes\": [\"openid\", \"pid\", \"example.read\", \"example.global_read\"]","WWW-Authenticate: Bearer error=\"insufficient_scope\", scope=\"example.global_read\""],"workaround":null,"fixed_in":null,"score":60.1,"first_seen":"2026-10-19T18:38:33Z"},{"id":"anthropics/claude-code#96662","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":96662,"title":"[BUG] Auto Mode classifier false positives in legitimate multi-agent repo workflow after recent Claude Code update","url":"https://github.com/anthropics/claude-code/issues/96662","state":"open","state_reason":null,"labels":["bug","platform:windows","area:permissions"],"created":"2026-09-24T09:30:49Z","updated":"2026-10-13T13:46:11Z","closed":null,"reactions":3,"thumbs_up":3,"comments":4,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update","crash/hang","auth/session"],"cosmetic_ui":false,"summary":"After updating Claude Code to the latest version in order to use Opus 5.5, Auto Mode started blocking legitimate, user-directed actions in an existing multi-session repository workflow. This workflow had been used repeatedly before the…","versions":[],"error_snippets":["After updating Claude Code to the latest version in order to use Opus 5.5, Auto Mode started blocking legitimate, user-directed actions in an existing multi-session repository workflow.","I have encountered three different false positives:","I explicitly asked Claude Code to use the configuration/update skill to add a permission rule.","Auto Mode blocked the requested action as `[Self-Modification]`, even though the permission change had been explicitly requested by the user.","Publishing this normal audit/protocol entry was blocked as `[Auto-Mode Bypass]`."],"workaround":{"text":"Disable Auto Mode for the affected sessions. - Use `acceptEdits`. - Add explicit permission allow rules for the required commands.","source":"issue body","url":"https://github.com/anthropics/claude-code/issues/96662"},"fixed_in":null,"score":60.0,"first_seen":"2026-10-23T09:23:24Z"},{"id":"Aider-AI/aider#5882","harness":"aider","harness_name":"Aider","repo":"Aider-AI/aider","number":5882,"title":"aider-install installed but not found","url":"https://github.com/Aider-AI/aider/issues/5882","state":"open","state_reason":null,"labels":[],"created":"2026-10-19T06:59:34Z","updated":"2026-10-22T17:28:17Z","closed":null,"reactions":1,"thumbs_up":1,"comments":1,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"Installed python from Microsoft Store using Python Install Manager. Installed aider-install according to instructions: python -m aider-install But then aider-install is not found. Pip says that it is installed: Package Version…","versions":["0.2.0"],"error_snippets":["But then aider-install is not found."],"workaround":null,"fixed_in":null,"score":59.6,"first_seen":"2026-10-19T18:38:33Z"},{"id":"kirodotdev/Kiro#11990","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":11990,"title":"kiro-cli macOS app: \"Shell integrations ✘\" keeps coming back — Kiro and Docker Desktop fight over the top of ~/.zprofile and ~/.profile, and every Kiro rewrite adds blank lines","url":"https://github.com/kirodotdev/Kiro/issues/11990","state":"open","state_reason":null,"labels":["os: mac","cli"],"created":"2026-10-17T14:18:19Z","updated":"2026-10-19T15:24:06Z","closed":null,"reactions":0,"thumbs_up":0,"comments":4,"severity":"medium","severity_reasons":["class:self-host/headless"],"topic":"self-host/headless","classes":["self-host/headless","install/update"],"cosmetic_ui":false,"summary":"macOS 26.6.2 (25G83), arm64 kiro-cli 2.32.0 (macOS app Kiro CLI.app, dashboard bundle dated 2026-10-15). Docker Desktop installed and running. The macOS app's Help & support → Automated checks keeps reporting Shell integrations ✘ with an…","versions":["2.32.0"],"error_snippets":["export PATH=\"$PATH:…\"","should fix everything it flags,** including","The cause is a tug-of-war over `~/.zprofile` and `~/.profile` between Kiro and Docker Desktop. Kiro also leaves extra blank lines in those files every time it rewrites them.","Docker Desktop's PATH block (`# The following lines were added by Docker Desktop to add commands to your PATH.`) is present in exactly `.zprofile` and `.profile`, and in neither `.zshrc` nor `.bashrc`. Docker Desktop is ","After each rewrite the Kiro pre block is first again, and by the next snapshot Docker's block is back on top. I'm inferring that Docker Desktop re-inserts its block when it starts. I haven't traced Docker's side."],"workaround":null,"fixed_in":null,"score":59.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aaif-goose/goose#13128","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":13128,"title":"Goose APP UPDATE BUG","url":"https://github.com/aaif-goose/goose/issues/13128","state":"open","state_reason":null,"labels":[],"created":"2026-10-25T04:01:14Z","updated":"2026-10-25T23:56:59Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"Describe the bug 💡 Before filing, please check common issues: https://goose-docs.ai/docs/troubleshooting 📦 To help us debug faster, attach your diagnostics zip if possible. 👉 How to capture it…","versions":["1.0.2"],"error_snippets":["4. See error","A clear and concise description of what you expected to happen.","If applicable, add screenshots to help explain your problem."],"workaround":null,"fixed_in":null,"score":59.5,"first_seen":"2026-10-25T09:11:37Z"},{"id":"openclaw/openclaw#181490","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":181490,"title":"[Bug]: macOS 2026.10.5 to 2026.10.6 update stalls in candidate Doctor with waiting SQLite worker","url":"https://github.com/openclaw/openclaw/issues/181490","state":"closed","state_reason":"completed","labels":["P2","clawsweeper:no-new-fix-pr","clawsweeper:needs-maintainer-review","issue-rating: 🦪 silver shellfish","impact:ux-friction"],"created":"2026-10-23T08:15:16Z","updated":"2026-10-24T06:02:34Z","closed":"2026-10-24T06:02:34Z","reactions":2,"thumbs_up":0,"comments":8,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update","always-on/daemon","crash/hang","network"],"cosmetic_ui":false,"summary":"Crash (process/app exits or hangs) On a macOS npm-global installation, updating Gateway 2026.10.5 to 2026.10.6 stalls in candidate Doctor before activation, with a waiting Doctor process and a waiting SQLite read-only worker; interrupting…","versions":["2026.10.6","2026.10.5"],"error_snippets":["Doctor event loop waiting","08:00:28 run status: failed / finished / interrupted","08:09:09 CLI exit code: 130","08:10:26 all upgrade processes confirmed exited","Earlier attempts:"],"workaround":null,"fixed_in":null,"score":59.4,"first_seen":"2026-10-23T09:23:24Z"},{"id":"openclaw/openclaw#169997","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":169997,"title":"Update failure: validating (2026.9.6)","url":"https://github.com/openclaw/openclaw/issues/169997","state":"open","state_reason":null,"labels":["clawsweeper:no-new-fix-pr","clawsweeper:needs-maintainer-review","P0","issue-rating: 🦪 silver shellfish","impact:ux-release-blocker"],"created":"2026-10-12T16:06:39Z","updated":"2026-10-14T02:58:36Z","closed":null,"reactions":2,"thumbs_up":0,"comments":3,"severity":"critical","severity_reasons":["label:p0"],"topic":"install/update","classes":["install/update","always-on/daemon"],"cosmetic_ui":false,"summary":"This report was explicitly reviewed and confirmed in OpenClaw. OpenClaw version: 2026.9.6 Node version: 24.18.0 Update action: Gateway RPC: update.run (control-ui) Installation method: npm-global Update target: 2026.10.2 Failed phase…","versions":["2026.9.6","24.18.0","2026.10.2"],"error_snippets":["This report was explicitly reviewed and confirmed in OpenClaw."],"workaround":null,"fixed_in":null,"score":59.3,"first_seen":"2026-10-18T04:33:05Z"},{"id":"github/copilot-cli#5223","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":5223,"title":"Interactive autopilot stops after 600s while a background subagent is still running (regression in 1.0.102)","url":"https://github.com/github/copilot-cli/issues/5223","state":"open","state_reason":null,"labels":["area:agents"],"created":"2026-10-21T21:43:57Z","updated":"2026-10-23T23:08:36Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"high","severity_reasons":["title:breakage"],"topic":"always-on/daemon","classes":["always-on/daemon","install/update","crash/hang","network"],"cosmetic_ui":false,"summary":"Since 1.0.102, interactive autopilot stops whenever a background subagent runs for more than 600 seconds. The subagent is still working normally when this happens. The CLI shows: A common pattern is for the main agent to start a background…","versions":["1.0.102","1.0.101","1.0.80"],"error_snippets":["COPILOT_TASK_WAIT_TIMEOUT_SECONDS","autopilot_background_timeout","export COPILOT_TASK_WAIT_TIMEOUT_SECONDS=3600","COPILOT_TASK_WAIT_TIMEOUT_SECONDS=0","> Autopilot stopped after 600s waiting for background tasks. Set COPILOT_TASK_WAIT_TIMEOUT_SECONDS to change the limit."],"workaround":{"text":"In 1.0.102 the native runtime adds new autopilot continuation logic, and the stop message and `autopilot_background_timeout` warning first appear there. - Workaround: `export COPILOT_TASK_WAIT_TIMEOUT_SECONDS=3600` (or higher) before launching. - The limit can't be turned off.","source":"issue body","url":"https://github.com/github/copilot-cli/issues/5223"},"fixed_in":null,"score":59.3,"first_seen":"2026-10-24T08:54:58Z"},{"id":"github/copilot-cli#1838","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":1838,"title":"Bug: Copilot CLI hangs in Nix/direnv environments due to subprocess I/O deadlock","url":"https://github.com/github/copilot-cli/issues/1838","state":"closed","state_reason":"completed","labels":["area:tools"],"created":"2026-03-05T06:17:46Z","updated":"2026-09-28T10:29:28Z","closed":"2026-09-28T10:29:28Z","reactions":12,"thumbs_up":12,"comments":7,"severity":"high","severity_reasons":["title:breakage"],"topic":"crash/hang","classes":["crash/hang","network"],"cosmetic_ui":false,"summary":"Copilot CLI v0.0.421 hangs indefinitely when launched from directories with Nix flake-based development environments managed by direnv. The bash tool fails to execute any commands, timing out with Invalid shell ID errors. Analysis of the…","versions":["0.0.421"],"error_snippets":["# Create a minimal flake-based dev environment","# Create flake.nix with direnv-managed environment","flake-utils.lib.eachDefaultSystem (system:","# Create .envrc","cat > .envrc << 'ENVRC'"],"workaround":null,"fixed_in":null,"score":59.3,"first_seen":"2026-10-19T18:38:33Z"},{"id":"NousResearch/hermes-agent#112028","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":112028,"title":"Support multi-surface access to the same session (desktop + mobile dashboard)","url":"https://github.com/NousResearch/hermes-agent/issues/112028","state":"open","state_reason":null,"labels":["type/feature","comp/gateway","P2","needs-decision","sweeper:risk-session-state","area/sessions","comp/tui-gateway"],"created":"2026-09-15T15:34:47Z","updated":"2026-10-20T19:03:34Z","closed":null,"reactions":2,"thumbs_up":0,"comments":7,"severity":"medium","severity_reasons":["class:always-on/daemon"],"topic":"always-on/daemon","classes":["always-on/daemon","self-host/headless"],"cosmetic_ui":false,"summary":"Hermes Agent: v0.21.3 Server: macOS (Mac mini) Surfaces: Desktop app + mobile dashboard When a session is open on the Desktop app (surface=desktop), a second surface (mobile dashboard, surface=tui) cannot submit messages to that same…","versions":["0.21.3"],"error_snippets":[") cannot submit messages to that same session — the server rejects it with","When a session is open on the Desktop app (surface=`desktop`), a second surface (mobile dashboard, surface=`tui`) cannot submit messages to that same session — the server rejects it with `SESSION_NOT_OWNED` (\"Session … a","Root cause found during investigation: `track_liveness` in `tui_gateway/session_lifecycle.py` is enabled only for `surface == \"desktop\"`, so a Desktop-opened session holds its lease indefinitely; `active_sessions.try_acq","3. At minimum: auto-release the lease when the owner surface is idle/unfocused, and return a distinct error code clients can surface to the user."],"workaround":null,"fixed_in":null,"score":59.2,"first_seen":"2026-10-21T09:24:21Z"},{"id":"anthropics/claude-code#97954","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":97954,"title":"[BUG] Cowork (Windows): tools/MCP connectors become unavailable as soon as voice mode is activated","url":"https://github.com/anthropics/claude-code/issues/97954","state":"open","state_reason":null,"labels":["bug","platform:windows","area:cowork"],"created":"2026-09-28T18:58:36Z","updated":"2026-10-09T05:24:31Z","closed":null,"reactions":3,"thumbs_up":3,"comments":4,"severity":"medium","severity_reasons":["class:mcp/tools"],"topic":"mcp/tools","classes":["mcp/tools","self-host/headless"],"cosmetic_ui":false,"summary":"In a Cowork conversation on Claude Desktop for Windows, tools work normally in text mode (including a local MCP server — Obsidian — that reads/writes my vault). As soon as I switch the same conversation to voice mode, the tools become…","versions":["2.9939.2"],"error_snippets":["No visible error message in the UI. Claude simply reports that the tools are not available once voice mode is on.","Voice mode in Cowork should have access to the same tools and connectors (including local MCP servers, especifically Obsidian in my case) as text mode in the same conversation, so Claude can keep calling them while I tal","Claude Desktop 2.9939.2 (Windows x64, Electron 44.4.3) — Cowork, not the Claude Code CLI"],"workaround":null,"fixed_in":null,"score":59.2,"first_seen":"2026-10-19T18:38:33Z"},{"id":"NousResearch/hermes-agent#122326","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":122326,"title":"tools.lazy_deps.install_specs shim starts a full update takeover from hermes serve/gateway; Hindsight plugin turns it into a Desktop rebuild loop","url":"https://github.com/NousResearch/hermes-agent/issues/122326","state":"closed","state_reason":"completed","labels":["type/bug","comp/cli","comp/tools","comp/plugins","tool/memory","P3","sweeper:risk-compatibility","area/install-update"],"created":"2026-09-25T05:20:16Z","updated":"2026-10-17T10:20:25Z","closed":"2026-10-06T13:12:39Z","reactions":2,"thumbs_up":2,"comments":10,"severity":"medium","severity_reasons":["class:always-on/daemon","class:install/update"],"topic":"always-on/daemon","classes":["always-on/daemon","install/update","mcp/tools","self-host/headless"],"cosmetic_ui":false,"summary":"tools.lazydeps.installspecs is now a shim that calls hermescli.oldupdater.stopforrelaunch(). That function starts the full historical update takeover (updatetakeover.py → updatefinish.py), which syncs PM and rebuilds Node deps, the TUI…","versions":["1.0.1","0.10.1"],"error_snippets":["_ensure_local_runtime (hindsight\\embedded.py:133)","is_available (hindsight\\__init__.py:446)","{\"restart_update\": false, \"plan\": null, \"argv\": [\"...\\\\.hermes\\\\bin\\\\hermes\", \"serve\", \"--host\", \"127.0.0.1\", \"--port\", \"0\"], ...}","hindsight-embed","appears at 00:43, 00:44, 00:54, 00:56 and 01:08. Each timestamp matches the start of a Desktop rebuild in"],"workaround":{"text":"Setting `memory.provider: ''` stops the loop. After Desktop relaunched, agent init finished normally and no takeover started. Hindsight memory is unavailable while this setting is in place.","source":"issue body","url":"https://github.com/NousResearch/hermes-agent/issues/122326"},"fixed_in":null,"score":59.1,"first_seen":"2026-10-18T04:32:26Z"},{"id":"google-gemini/gemini-cli#29830","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":29830,"title":"[Crash]: FATAL ERROR: Reached heap limit Allocation failed - JavaScript heap out of memory (V8 Heap Limit Exceeded at ~8.3GB)","url":"https://github.com/google-gemini/gemini-cli/issues/29830","state":"open","state_reason":null,"labels":["area/core","status/bot-triaged","effort/large"],"created":"2026-10-19T17:08:30Z","updated":"2026-10-20T03:01:57Z","closed":null,"reactions":0,"thumbs_up":0,"comments":4,"severity":"high","severity_reasons":["title:breakage"],"topic":"crash/hang","classes":["crash/hang"],"cosmetic_ui":false,"summary":"The Gemini CLI process stopped with a fatal error after approximately 9.3 hours of one interactive session. The workspace had Playwright reports, Allure results, and JSON files. The heap increased to more than 8.29 GB. The V8 garbage…","versions":["0.64.0","24.10.0","5.1","9.33"],"error_snippets":["FATAL ERROR: Reached heap limit Allocation failed - JavaScript heap out of memory","[19548:000001CC31055000] 33617281 ms: Scavenge 7927.0 (7955.5) -> 7922.4 (7954.7) MB, pooled: 0 MB, 0.73 / 0.00 ms (average mu = 0.995, current mu = 0.996) allocation failure;","[19548:000001CC31055000] 33618067 ms: Mark-Compact 8292.0 (8322.1) -> 8285.2 (8310.0) MB, pooled: 11 MB, 192.84 / 0.01 ms (average mu = 0.988, current mu = 0.919) allocation failure; scavenge might not succeed","1: 00007FF6CC237DBF node::OnFatalError+1343","2: 00007FF6CCE7B017 v8::Function::NewInstance+423"],"workaround":null,"fixed_in":null,"score":59.1,"first_seen":"2026-10-19T18:38:33Z"},{"id":"google-gemini/gemini-cli#29772","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":29772,"title":"Persistent 429 RESOURCE_EXHAUSTED on Gemini Code Assist Standard","url":"https://github.com/google-gemini/gemini-cli/issues/29772","state":"open","state_reason":null,"labels":["status/need-triage","area/platform"],"created":"2026-10-15T20:58:25Z","updated":"2026-10-19T18:27:24Z","closed":null,"reactions":0,"thumbs_up":0,"comments":5,"severity":"medium","severity_reasons":["class:network"],"topic":"network","classes":["network"],"cosmetic_ui":false,"summary":"I purchased a Gemini Code Assist Standard subscription and successfully assigned a license to my account. Gemini CLI correctly recognizes my tier as \"Gemini Code Assist Standard\". However, after assigning the license, every request returns…","versions":[],"error_snippets":["to enter the interactive CLI, then run the","However, after assigning the license, every request returns RESOURCE_EXHAUSTED, no matter how long I wait. Gemini CLI automatically retries, but it remains the same no matter how long it retries, and I am still unable to","As a user who has already been assigned a Gemini Code Assist Standard license, I expected to be able to use Gemini CLI normally after the license was assigned.","Since the CLI already correctly recognizes my tier as Standard, requests should succeed. If the account has been restricted or flagged, I would also appreciate a clear explanation and information on whether an unban can ","Run `gemini` to enter the interactive CLI, then run the `/about` command."],"workaround":null,"fixed_in":null,"score":59.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aaif-goose/goose#11323","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":11323,"title":"ACP desktop drops CallToolResult._meta before MCP App notification","url":"https://github.com/aaif-goose/goose/issues/11323","state":"open","state_reason":null,"labels":[],"created":"2026-08-18T13:44:38Z","updated":"2026-10-19T17:06:34Z","closed":null,"reactions":1,"thumbs_up":1,"comments":7,"severity":"medium","severity_reasons":["class:mcp/tools"],"topic":"mcp/tools","classes":["mcp/tools","self-host/headless","install/update"],"cosmetic_ui":false,"summary":"Describe the bug Goose Desktop's ACP chat path drops the MCP server's non-routing CallToolResult.meta before delivering ui/notifications/tool-result to an MCP App. The app receives content and structuredContent, but meta is reconstructed…","versions":["1.46.0"],"error_snippets":["_meta: { \"example.com/canary\": { present: true } },","_meta[\"example.com/canary\"]","This has the same externally visible symptom as #7435/#7476, but IIUC occurs at a different location. The prior fix covered the legacy desktop path, the current ACP path performs two lossy reconstructions.","4. Observe that `structuredContent.present` is available, but `_meta[\"example.com/canary\"]` is absent. `_meta` contains only Goose-derived app routing fields.","A namespaced canary assertion fails independently at each boundary."],"workaround":null,"fixed_in":null,"score":59.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"omacom/omarchy#12944","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":12944,"title":"Cloned menu plugin loses its scoped shell API during shell startup - installed apps vanish from menu and search","url":"https://github.com/omacom/omarchy/issues/12944","state":"open","state_reason":null,"labels":["bug"],"created":"2026-09-22T20:06:24Z","updated":"2026-10-13T02:31:09Z","closed":null,"reactions":1,"thumbs_up":1,"comments":6,"severity":"medium","severity_reasons":["class:install/update","class:mcp/tools"],"topic":"install/update","classes":["install/update","mcp/tools"],"cosmetic_ui":false,"summary":"AMD Ryzen AI 5 340 w/ Radeon 840M (Framework), Omarchy 4.0.4-1, Quickshell 0.3.1-1, Hyprland 0.56.2, kernel 7.2.5-3-omarchy, Wayland. After cloning the menu plugin with the documented flow (omarchy plugin clone omarchy.menu), installed…","versions":["4.0.4-1","0.3.1-1","0.56.2","4.0.4"],"error_snippets":["Component.onCompleted: shell = null (expected)","Timer 500ms tick 1: shell = null (already revoked, stays null)","mergeAppRows: appLibrary = null (apps silently skipped)","4. No app rows appear; `omarchy menu summon apps` shows an empty submenu","Evidence from temporary instrumentation in the cloned `Menu.qml`:"],"workaround":null,"fixed_in":null,"score":58.9,"first_seen":"2026-10-25T09:11:37Z"},{"id":"continuedev/continue#13417","harness":"continue","harness_name":"Continue","repo":"continuedev/continue","number":13417,"title":"fix(openai): Stream parser crashes on extra unknown JSON fields (extra_content / usage payload) in CLI (cn) & IDE","url":"https://github.com/continuedev/continue/issues/13417","state":"open","state_reason":null,"labels":[],"created":"2026-10-24T14:41:00Z","updated":"2026-10-24T15:03:21Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"high","severity_reasons":["title:breakage"],"topic":"crash/hang","classes":["crash/hang","network"],"cosmetic_ui":false,"summary":"I've tried finding an answer on the Continue docs site I'm not able to find an open issue that reports the same bug I've seen the troubleshooting guide on the Continue Docs When using an OpenAI-compatible provider whose SSE stream includes…","versions":["1.0.67","2025.2.4","1.0.0"],"error_snippets":["- Model: Google Gemini OpenAI-compatibility endpoint","field in **every** chunk (e.g. Google Gemini's OpenAI compatibility endpoint","is expected only in the **final** chunk, which has an empty","array. Gemini's OpenAI-compat endpoint, however, returns","in every chunk **alongside** non-empty"],"workaround":null,"fixed_in":null,"score":58.8,"first_seen":"2026-10-25T09:11:37Z"},{"id":"cline/cline#15344","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":15344,"title":"CLI shell tools inherit shared Hub startup environment instead of launching client","url":"https://github.com/cline/cline/issues/15344","state":"open","state_reason":null,"labels":[],"created":"2026-10-26T03:17:39Z","updated":"2026-10-26T03:17:43Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"low","severity_reasons":["low engagement"],"topic":"always-on/daemon","classes":["always-on/daemon","auth/session"],"cosmetic_ui":false,"summary":"Cline CLI 3.0.73 on macOS reuses an existing local Hub. A shell tool invoked by the model then sees the terminal context from whichever client started that Hub. Launch Cline in terminal A and leave its Hub running. In terminal B export a…","versions":["3.0.73"],"error_snippets":["Cline CLI 3.0.73 on macOS reuses an existing local Hub. A shell tool invoked by the model then sees the terminal context from whichever client started that Hub.","2. In terminal B export a fresh non-secret canary, launch the interactive CLI, and ask it to run a shell probe comparing that canary and terminal identifiers with terminal B.","3. Compare only presence/equality; do not print credentials or a complete environment.","A real model-invoked Python child ran under the pre-existing Hub. Terminal pane/tab/workspace identifiers differed from the new launching shell, TERM_PROGRAM was absent, and five fresh synthetic environment canaries were","Shell commands receive session-scoped launch context. Two clients sharing the Hub must not inherit each other’s terminal identifiers, and no request may mutate the Hub’s global environment."],"workaround":null,"fixed_in":null,"score":58.7,"first_seen":"2026-10-26T10:12:02Z"},{"id":"NousResearch/hermes-agent#9153","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":9153,"title":"Docker image missing 'dashboard' command (web UI not available in latest image)","url":"https://github.com/NousResearch/hermes-agent/issues/9153","state":"closed","state_reason":"completed","labels":["type/bug","comp/cli","P2","area/install-update"],"created":"2026-04-13T17:19:01Z","updated":"2026-10-17T13:25:19Z","closed":"2026-05-07T13:16:25Z","reactions":11,"thumbs_up":11,"comments":0,"severity":"medium","severity_reasons":["class:self-host/headless","class:install/update"],"topic":"self-host/headless","classes":["self-host/headless","install/update"],"cosmetic_ui":false,"summary":"The nousresearch/hermes-agent:latest Docker image on Docker Hub does not include the dashboard (or web) subcommand, making it impossible to run the web UI in containerized environments like Coolify. The dashboard command should be…","versions":[],"error_snippets":["hermes: error: argument command: invalid choice: 'dashboard' (choose from chat, model, gateway, setup, ...)","command exists in the current","The `nousresearch/hermes-agent:latest` Docker image on Docker Hub does not include the `dashboard` (or `web`) subcommand, making it impossible to run the web UI in containerized environments like Coolify."],"workaround":null,"fixed_in":null,"score":58.6,"first_seen":"2026-10-18T04:29:32Z"},{"id":"github/copilot-cli#5088","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":5088,"title":"Windows: CLI crashes (0xc0000005 in msalruntime.dll) on first Entra account-broker sign-in for MCP server - interactive WAM called with no parent window","url":"https://github.com/github/copilot-cli/issues/5088","state":"open","state_reason":null,"labels":["area:authentication","area:platform-windows","area:mcp"],"created":"2026-10-08T17:28:39Z","updated":"2026-10-12T09:51:38Z","closed":null,"reactions":1,"thumbs_up":1,"comments":1,"severity":"high","severity_reasons":["title:breakage"],"topic":"self-host/headless","classes":["self-host/headless","crash/hang","auth/session","mcp/tools","install/update"],"cosmetic_ui":false,"summary":"On Windows, the first Microsoft Entra account broker (WAM) sign-in for a remote MCP server on the built-in Entra fast-path list (here https://sentinel.microsoft.com/mcp/data-exploration) hard-crashes the CLI process with an access…","versions":["1.0.91","1.0.94-4","2.0"],"error_snippets":["\"Microsoft Sentinel Data Exploration\": { \"type\": \"http\", \"url\": \"https://sentinel.microsoft.com/mcp/data-exploration\", \"tools\": [\"*\"] }","[ERROR] Connecting to Microsoft Sentinel Data Exploration...","[ERROR] Signing in with your Microsoft Entra account (account broker)...","Exception code: 0xc0000005","https://sentinel.microsoft.com/mcp/data-exploration"],"workaround":null,"fixed_in":null,"score":58.6,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aaif-goose/goose#12909","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":12909,"title":"Document the Windows installer for goose Desktop","url":"https://github.com/aaif-goose/goose/issues/12909","state":"open","state_reason":null,"labels":[],"created":"2026-10-15T05:39:03Z","updated":"2026-10-22T09:21:50Z","closed":null,"reactions":0,"thumbs_up":0,"comments":4,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"What problem would this solve? PR #12869 (closes #11987) added a signed, per-user NSIS installer (Goose-win32-x64-setup.exe) for goose Desktop on Windows. From the next release it will be published alongside the existing ZIP, but the…","versions":[],"error_snippets":[],"workaround":null,"fixed_in":null,"score":58.1,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openclaw/openclaw#158990","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":158990,"title":"[Bug]: Gateway temp log truncated on every CLI/gateway start destroys crash forensics (Windows)","url":"https://github.com/openclaw/openclaw/issues/158990","state":"open","state_reason":null,"labels":["stale","P2","impact:data-loss","issue-rating: 🦪 silver shellfish"],"created":"2026-09-26T15:42:58Z","updated":"2026-10-24T21:35:31Z","closed":null,"reactions":2,"thumbs_up":0,"comments":2,"severity":"critical","severity_reasons":["label:data-loss"],"topic":"always-on/daemon","classes":["always-on/daemon","crash/hang"],"cosmetic_ui":false,"summary":"On Windows, the daily gateway log at %TEMP%\\openclaw\\openclaw- .log is truncated and rebuilt every time a CLI command or the gateway itself starts. After a gateway crash, the first thing that destroys the crash evidence is the operator's…","versions":["2026.9.6"],"error_snippets":["> **Disclosure: this issue was written by an AI coding agent** on behalf of the operator. All measurements below were produced directly from a running 2026.9.6 (Windows) instance during an incident investigation on 2026-","On Windows, the daily gateway log at `%TEMP%\\openclaw\\openclaw- .log` is **truncated and rebuilt every time a CLI command or the gateway itself starts**. After a gateway crash, the first thing that destroys the crash evi","Consequence: post-crash forensics on Windows are only possible if the operator copies the log **before** running any OpenClaw command. That is backwards — status / doctor / restart is the first thing an operator runs, an"],"workaround":null,"fixed_in":null,"score":58.0,"first_seen":"2026-10-25T09:11:37Z"},{"id":"kirodotdev/Kiro#11958","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":11958,"title":"\">50 MCP tools enabled\" warning still shows in the IDE (1.3) after Tool Search was introduced","url":"https://github.com/kirodotdev/Kiro/issues/11958","state":"open","state_reason":null,"labels":["os: windows","mcp","ide","usability"],"created":"2026-10-15T12:28:42Z","updated":"2026-10-15T14:29:37Z","closed":null,"reactions":1,"thumbs_up":1,"comments":2,"severity":"medium","severity_reasons":["class:mcp/tools"],"topic":"mcp/tools","classes":["mcp/tools","self-host/headless","network"],"cosmetic_ui":false,"summary":"I see that on the Kiro IDE 1.3.0 - where Tool Search was recently added and is on by default - I still get the warning: You have N MCP tools enabled. We recommend disabling servers or tools to keep this below 50. At the same time, the IDE…","versions":["1.3","1.3.0"],"error_snippets":["You have N MCP tools enabled. We recommend disabling servers or tools to keep this below 50.","`You have N MCP tools enabled. We recommend disabling servers or tools to keep this below 50.`","At the same time, the IDE context-usage meter reports MCP tools at 0% in a fresh session, and tool definitions appear to load on demand rather than up front — which is what I'd expect Tool Search to do.","2. Configure MCP servers totalling more than 50 enabled tools.","4. Observe the \"You have N MCP tools enabled… keep this below 50\" warning."],"workaround":null,"fixed_in":null,"score":58.0,"first_seen":"2026-10-22T09:27:06Z"},{"id":"aaif-goose/goose#13053","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":13053,"title":"[Bug] goose doctor panics in sqlx-sqlite row.rs after upgrading 1.54.0 -> 1.55.0 (regression of #8719)","url":"https://github.com/aaif-goose/goose/issues/13053","state":"open","state_reason":null,"labels":[],"created":"2026-10-21T16:29:27Z","updated":"2026-10-23T12:42:08Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"high","severity_reasons":["title:breakage"],"topic":"install/update","classes":["install/update","crash/hang","always-on/daemon"],"cosmetic_ui":false,"summary":"Describe the bug After upgrading from 1.54.0 to 1.55.0, goose doctor panics immediately in a background sqlx worker thread: Exit code: 1. This looks like a regression of #8719 (fix: avoid sqlite session creation panic on legacy dbs, closed…","versions":["1.54.0","1.55.0","0.9.0","0.8.6"],"error_snippets":["thread 'sqlx-sqlite-worker-0' (35808) panicked at","note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace","error: Could not create session: no rows returned by a query that expected to return at least one row","fix: avoid sqlite session creation panic on legacy dbs","no rows returned by a query that expected to return at least one row"],"workaround":{"text":"Roll back to 1.54.0, which works against the same database:","source":"issue body","url":"https://github.com/aaif-goose/goose/issues/13053"},"fixed_in":null,"score":58.0,"first_seen":"2026-10-22T09:27:06Z"},{"id":"omacom/omarchy#15831","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":15831,"title":"linux-omarchy 7.2.8-9: webcam broken on XPS 13 9340 (IPU6), HIMX1092 ipu-bridge entry blocks ISYS notifier without an hm1092 driver","url":"https://github.com/omacom/omarchy/issues/15831","state":"open","state_reason":null,"labels":["bug","compatibility"],"created":"2026-10-18T11:36:53Z","updated":"2026-10-23T03:35:12Z","closed":null,"reactions":1,"thumbs_up":1,"comments":2,"severity":"high","severity_reasons":["title:breakage"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"Omarchy 4.0.5-1, linux-omarchy 7.2.8-9 (regression from 7.2.5-3) Dell XPS 13 9340, BIOS 1.29.0, Meteor Lake, IPU6-v4, Intel IVSC Webcam: OV02C10 (OVTI02C1, \\SB.PC00.LNK1) via IVSC → IPU6 CSI2 4 IR camera: Himax HM1092 (HIMX1092…","versions":["7.2.8-9","4.0.5-1","7.2.5-3","6.18"],"error_snippets":["node. Only 7 exist (6× CSI2 + IVSC CSI). PipeWire instead exposes ~48 raw ipu6 nodes, which fail with","entry (backport of mainline","ERROR V4L2 v4l2_device.cpp:92 'ov02c10 17-0036': Failed to open V4L2 device '': No such file or directory","ERROR CameraSensor camera_sensor.cpp:470 Failed to create sensor for 'ov02c10 17-0036: -2","ERROR SimplePipeline simple.cpp:1930 No valid pipeline for sensor 'ov02c10 17-0036', skipping"],"workaround":null,"fixed_in":null,"score":57.9,"first_seen":"2026-10-23T09:23:24Z"},{"id":"NousResearch/hermes-agent#134381","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":134381,"title":"[Bug]: Update failed","url":"https://github.com/NousResearch/hermes-agent/issues/134381","state":"open","state_reason":null,"labels":["type/bug","comp/cli","P2","needs-repro","sweeper:risk-compatibility","comp/desktop","area/install-update"],"created":"2026-10-07T06:43:34Z","updated":"2026-10-17T07:50:31Z","closed":null,"reactions":4,"thumbs_up":4,"comments":0,"severity":"high","severity_reasons":["title:breakage"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"Hi i tried to update the hermes client on MacOS but it failed. Error message is only Update failed so no info there :). Hermes Agent v0.21.5+8076.g85db7c3 (2026.9.24) · upstream 85db7c3a Install directory…","versions":["0.21.5+8076.g85db7c3","2026.9.24","2.24.0"],"error_snippets":["Hi i tried to update the hermes client on MacOS but it failed. Error message is only Update failed so no info there :).","Run Hermes Desktop App and press update -> failure"],"workaround":null,"fixed_in":null,"score":57.8,"first_seen":"2026-10-18T04:32:26Z"},{"id":"NousResearch/hermes-agent#111942","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":111942,"title":"[Bug]: CLI crashes on startup with NameError: name 'file_signature' is not defined (cli_tui_mixin.py, regression from #111408)","url":"https://github.com/NousResearch/hermes-agent/issues/111942","state":"closed","state_reason":"completed","labels":["type/bug","comp/cli","area/config","P0"],"created":"2026-09-15T14:05:17Z","updated":"2026-09-16T11:21:20Z","closed":"2026-09-15T15:17:08Z","reactions":5,"thumbs_up":4,"comments":12,"severity":"critical","severity_reasons":["label:p0"],"topic":"install/update","classes":["install/update","crash/hang"],"cosmetic_ui":false,"summary":"After updating to the latest main (24fd22b94d, v0.21.3 / 2026.9.14), the interactive CLI crashes immediately on startup with: This is a regression from 182ec5c28d (#111408, \"widen the remaining config.yaml stat caches to filesignature\")…","versions":["0.21.3","2026.9.14","3.11.16"],"error_snippets":["NameError: name 'file_signature' is not defined","Traceback (most recent call last):","sys.exit(main())","self._config_sig: tuple | None = file_signature(_cfg_path.stat()) if _cfg_path.exists() else None","to an empty tempdir, so"],"workaround":null,"fixed_in":null,"score":57.7,"first_seen":"2026-10-18T04:29:32Z"},{"id":"openclaw/openclaw#164113","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":164113,"title":"[Bug]: update fails at updater-runtime-retention with FICLONE EPERM inside an unprivileged LXC container (seccomp blocks ioctl)","url":"https://github.com/openclaw/openclaw/issues/164113","state":"closed","state_reason":"completed","labels":["no-stale","clawsweeper:fix-shape-clear","clawsweeper:queueable-fix","clawsweeper:source-repro","P0","issue-rating: 🦞 diamond lobster","maturity:stable","impact:ux-release-blocker"],"created":"2026-10-03T07:04:39Z","updated":"2026-10-16T05:34:06Z","closed":"2026-10-05T02:20:43Z","reactions":2,"thumbs_up":1,"comments":13,"severity":"critical","severity_reasons":["label:p0"],"topic":"self-host/headless","classes":["self-host/headless","install/update","always-on/daemon"],"cosmetic_ui":false,"summary":"Behavior bug (incorrect output/state without crash) On an npm-global 2026.9.7 Gateway running inside an unprivileged Proxmox LXC container, openclaw update to the next candidate (2026.9.8) fails deterministically in phase validating at…","versions":["2026.9.7","2026.9.8","7.0.0","2.4.4-pve1"],"error_snippets":["Failed: validating","Failed: updater-runtime-retention - Exit code: 1","native file copy failed | FICLONE: Operation not permitted (os error 1) | EPERM","Failing check updater-runtime-retention (Error)","ioctl errno 1 [1,0x9400,SCMP_CMP_MASKED_EQ,0xff00]"],"workaround":null,"fixed_in":null,"score":57.7,"first_seen":"2026-10-18T04:33:05Z"},{"id":"openclaw/openclaw#159783","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":159783,"title":"[Bug]: 2026.9.6 Gateway heap grows ~45-100 MiB/h at rest on Linux systemd --user until V8 OOM","url":"https://github.com/openclaw/openclaw/issues/159783","state":"open","state_reason":null,"labels":["P1","clawsweeper:no-new-fix-pr","clawsweeper:needs-maintainer-review","impact:crash-loop","issue-rating: 🦪 silver shellfish"],"created":"2026-09-27T16:49:30Z","updated":"2026-10-18T03:40:23Z","closed":null,"reactions":2,"thumbs_up":1,"comments":3,"severity":"critical","severity_reasons":["label:crash-loop"],"topic":"always-on/daemon","classes":["always-on/daemon","crash/hang","install/update"],"cosmetic_ui":false,"summary":"On 2026.9.6 the Gateway's main-isolate heap grows steadily at rest, about 45–100 MiB/h, until it reaches Node's default V8 ceiling (~4.2 GiB on this host) and aborts with Ineffective mark-compacts near heap limit. It happened once…","versions":["2026.9.6","4.2","4035.1","4079.5"],"error_snippets":["FATAL ERROR: Ineffective mark-compacts near heap limit Allocation failed - JavaScript heap out of memory","`--max-old-space-size=10240` as a direct `node` argument in the service's `ExecStart` (not `NODE_OPTIONS`). That buys days; it does not stop the growth.","If a retainer path is needed, we can take a heap snapshot on our side and share an **aggregated** summary: constructor names, retained sizes and retainer chains, with no string contents. Raw snapshots would contain priva"],"workaround":{"text":"`--max-old-space-size=10240` as a direct `node` argument in the service's `ExecStart` (not `NODE_OPTIONS`). That buys days; it does not stop the growth.","source":"issue body","url":"https://github.com/openclaw/openclaw/issues/159783"},"fixed_in":null,"score":57.5,"first_seen":"2026-10-24T08:54:58Z"},{"id":"omacom/omarchy#2831","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":2831,"title":"mise python installation breaks AUR packages installations","url":"https://github.com/omacom/omarchy/issues/2831","state":"closed","state_reason":"completed","labels":["bug"],"created":"2025-10-25T01:52:15Z","updated":"2026-09-21T09:50:52Z","closed":"2026-09-21T09:50:52Z","reactions":10,"thumbs_up":10,"comments":9,"severity":"medium","severity_reasons":["engagement 19","class:install/update"],"topic":"install/update","classes":["install/update","network"],"cosmetic_ui":false,"summary":"11th Gen Intel(R) Core(TM) i5-1145G7 (8) @ 4.40 GHz, Intel Iris Xe Graphics @ 1.30 GHz [Integrated] Whenever a mise python version is installed, packages that depend on system python, like anki, will break builds and won't install: build…","versions":["4.40","1.30"],"error_snippets":["I would like to suggest pointing the python installation in the Menu > Install > Development to installing the `uv` python package manager directly instead of using `mise` to handle python. The current recommended workfl"],"workaround":null,"fixed_in":null,"score":57.4,"first_seen":"2026-10-19T18:38:33Z"},{"id":"google-gemini/gemini-cli#29747","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":29747,"title":"Body timeout is fixed at 300 s for API-key auth; a long silent tool call is dropped and replayed","url":"https://github.com/google-gemini/gemini-cli/issues/29747","state":"open","state_reason":null,"labels":["area/core","status/bot-triaged","effort/small"],"created":"2026-10-14T03:37:44Z","updated":"2026-10-14T15:34:31Z","closed":null,"reactions":0,"thumbs_up":0,"comments":2,"severity":"medium","severity_reasons":["class:auth/session","class:mcp/tools"],"topic":"auth/session","classes":["auth/session","mcp/tools","network","self-host/headless","install/update"],"cosmetic_ui":false,"summary":"With @google/gemini-cli 0.64.0 in headless mode (-p, --output-format stream-json), API-key auth and gemini-4-argon at thinkingLevel: HIGH, a turn in which the model writes one large tool-call argument (for example a whole document passed…","versions":["0.64.0"],"error_snippets":["exception TypeError: fetch failed sending request","(default 10). In our runs, two turns ended with","defaultBodyTimeout = 3e5","defaultHeadersTimeout = 6e4","updateGlobalFetchTimeouts"],"workaround":null,"fixed_in":null,"score":57.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aws/amazon-q-developer-cli#3684","harness":"amazon-q-cli","harness_name":"Amazon Q Developer CLI","repo":"aws/amazon-q-developer-cli","number":3684,"title":"bug: acp integration with intellij no longer works since update to kiro-cli 1.28.1","url":"https://github.com/aws/amazon-q-developer-cli/issues/3684","state":"closed","state_reason":"completed","labels":[],"created":"2026-03-25T10:15:56Z","updated":"2026-10-05T19:22:10Z","closed":"2026-10-05T19:22:10Z","reactions":11,"thumbs_up":11,"comments":3,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update","always-on/daemon","self-host/headless","auth/session","mcp/tools"],"cosmetic_ui":false,"summary":"I have run q doctor in the affected terminal session I have run q restart and replicated the issue again https://kiro.dev/docs/cli/acp/ acp integration used to work as per above instructions until update this week get an error in intellij…","versions":["1.28.1"],"error_snippets":["2026-03-25 09:55:29,182 [5034641] INFO - #c.i.m.l.a.a.p.AcpProcessHandlerService - Reusing existing process handler for agent 'Kiro' in chat 8de94113-1d33-43ac-8d91-e5641e0b1e5e","2026-03-25 09:55:31,869 [5037328] WARN - #c.i.m.l.a.a.c.a.StandardAuthHandler - Standard authentication failed for method kiro-login","com.agentclientprotocol.protocol.JsonRpcException: Method not found","at com.agentclientprotocol.protocol.Protocol.handleResponse(Protocol.kt:465)","at kotlinx.coroutines.scheduling.CoroutineScheduler$Worker.executeTask(CoroutineScheduler.kt:906)"],"workaround":null,"fixed_in":null,"score":57.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openclaw/openclaw#31331","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":31331,"title":"[Bug]: Docker Install + Sandbox can't workspaceAccess at all","url":"https://github.com/openclaw/openclaw/issues/31331","state":"closed","state_reason":"completed","labels":["bug","dedupe:parent","bug:behavior","P1","clawsweeper:no-new-fix-pr","clawsweeper:fix-shape-clear","clawsweeper:needs-maintainer-review","clawsweeper:needs-product-decision","clawsweeper:needs-live-repro","impact:session-state","impact:security","issue-rating: 🐚 platinum hermit","impact:ux-friction"],"created":"2026-03-02T05:47:25Z","updated":"2026-09-16T08:08:29Z","closed":"2026-09-14T01:22:49Z","reactions":5,"thumbs_up":4,"comments":9,"severity":"critical","severity_reasons":["label:security"],"topic":"self-host/headless","classes":["self-host/headless","install/update","always-on/daemon"],"cosmetic_ui":false,"summary":"Behavior bug (incorrect output/state without crash) When running the Gateway inside a Docker container and using Docker-outside-of-Docker for Sandboxing, /workspace can't binds properly. The Gateway containder attempts to mount the agent…","versions":["2026.2.26"],"error_snippets":["When running the Gateway inside a Docker container and using Docker-outside-of-Docker for Sandboxing, /workspace can't binds properly.","The Gateway containder attempts to mount the agent workspace using its **internal** path (e.g., /home/node/.openclaw/workspace), which does not exist on the **host** machine.","2. Enable sandbox","2. Sandbox default workdir can't be \"/workspace\" anymore, llm have to cd","4. It creates a emtpy /home/node path in host"],"workaround":{"text":"`\"sandbox\": { ; \"mode\": \"all\",`","source":"issue body","url":"https://github.com/openclaw/openclaw/issues/31331"},"fixed_in":null,"score":56.6,"first_seen":"2026-10-18T04:30:03Z"},{"id":"continuedev/continue#13372","harness":"continue","harness_name":"Continue","repo":"continuedev/continue","number":13372,"title":"Windows absolute paths are parsed as remote package slugs","url":"https://github.com/continuedev/continue/issues/13372","state":"open","state_reason":null,"labels":[],"created":"2026-10-13T16:26:33Z","updated":"2026-10-24T03:25:19Z","closed":null,"reactions":0,"thumbs_up":0,"comments":3,"severity":"medium","severity_reasons":["class:self-host/headless"],"topic":"self-host/headless","classes":["self-host/headless"],"cosmetic_ui":false,"summary":"Windows absolute paths are classified as remote package slugs by decodePackageIdentifier. On main (5522c6f), C:\\project\\rule.yaml, C:/project/rule.yaml, and \\\\server\\share\\rule.yaml all return uriType: \"slug\". The forward-slash drive path…","versions":[],"error_snippets":["On main (`5522c6f`), `C:\\project\\rule.yaml`, `C:/project/rule.yaml`, and `\\\\server\\share\\rule.yaml` all return `uriType: \"slug\"`. The forward-slash drive path even becomes owner `C:`, package `project`, version `rule.yam"],"workaround":null,"fixed_in":null,"score":56.6,"first_seen":"2026-10-19T18:38:33Z"},{"id":"anomalyco/opencode#58911","harness":"opencode","harness_name":"OpenCode","repo":"anomalyco/opencode","number":58911,"title":"Free usage limit does not reset after 24+ hours - stuck in infinite retry loop","url":"https://github.com/anomalyco/opencode/issues/58911","state":"closed","state_reason":"not_planned","labels":["pending close","triaging"],"created":"2026-10-25T06:53:34Z","updated":"2026-10-26T06:55:13Z","closed":"2026-10-26T06:55:12Z","reactions":2,"thumbs_up":2,"comments":2,"severity":"medium","severity_reasons":["class:install/update","class:crash/hang"],"topic":"install/update","classes":["install/update","crash/hang"],"cosmetic_ui":false,"summary":"I am using the free tier of OpenCode. I received the \"Free usage exceeded, subscribe to Go [retrying in XXXXXs]\" message. The problem is that the limit does NOT reset even after waiting for more than 24 hours and doing nothing in the app…","versions":[],"error_snippets":["I am using the free tier of OpenCode. I received the \"Free usage exceeded, subscribe to Go [retrying in XXXXXs]\" message.","The problem is that the limit does NOT reset even after waiting for more than 24 hours and doing nothing in the app. I have tried:","4. The \"Free usage exceeded\" error appears immediately again with a new ~18 hour timer.","The error persists indefinitely, making the free tier completely unusable."],"workaround":null,"fixed_in":null,"score":56.4,"first_seen":"2026-10-26T10:12:02Z"},{"id":"omacom/omarchy#15813","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":15813,"title":"Omarchy 4.0.5 / systemd 262: three TPM NvPCR services failed after update","url":"https://github.com/omacom/omarchy/issues/15813","state":"closed","state_reason":"completed","labels":["bug"],"created":"2026-10-18T08:39:41Z","updated":"2026-10-22T09:10:57Z","closed":"2026-10-22T09:10:57Z","reactions":3,"thumbs_up":3,"comments":4,"severity":"medium","severity_reasons":["class:always-on/daemon","class:install/update"],"topic":"always-on/daemon","classes":["always-on/daemon","install/update"],"cosmetic_ui":false,"summary":"Architecture: x8664 - CPU: Intel Pentium N3710 @ 1.60GHz - GPU: Intel Atom/Celeron/Pentium Processor x5-E8000/J3xxx/N3xxx Integrated Graphics Controller (rev 35) - Bootloader: Limine / UKI - Root filesystem: encrypted, Btrfs After updating…","versions":["4.0.5"],"error_snippets":["systemd-tpm2-setup-early.service","Failed to initialize NvPCR index: No such file or directory","Could not extend NvPCR: No such file or directory","Failed to start Early TPM SRK Setup.","Failed to start TPM NvPCR Product ID Measurement."],"workaround":null,"fixed_in":null,"score":56.3,"first_seen":"2026-10-19T18:38:33Z"},{"id":"kirodotdev/Kiro#11603","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":11603,"title":"Crew gateway claim-broker returns connections=0 for every session after update (identity_unattested on all MCP tools)","url":"https://github.com/kirodotdev/Kiro/issues/11603","state":"open","state_reason":null,"labels":["os: mac","mcp","auth","pending-maintainer-response"],"created":"2026-09-25T13:21:54Z","updated":"2026-10-16T18:01:25Z","closed":null,"reactions":0,"thumbs_up":0,"comments":5,"severity":"medium","severity_reasons":["class:always-on/daemon","class:install/update"],"topic":"always-on/daemon","classes":["always-on/daemon","install/update","auth/session","mcp/tools"],"cosmetic_ui":false,"summary":"Kiro Crew v0.7.1 After a KiroCrew auto-update installed today, every Crew MCP tool call is refused with identityunattested. The gateway’s session-claim handshake reports connections=0 on every claim-push — for dashboard sessions, scheduled…","versions":["0.7.1"],"error_snippets":["Scheduled crons cannot deliver (they run as sessions and hit the same wall). Observed live: cron:d6a67432 failed the claim at 09:00:15; the Morning Pulse cron cron:8ed32469 failed at ~08:10.","Live gateway PID 80445. Every claim-push returns claim-noop / updated=0 / connections=0, across all session kinds:","grep -c ‘connections=[1-9]’ over the whole log = 0 (no claim EVER bound a connection on this gateway generation).","Followed by, on each tool call:","Observe identity_unattested, and connections=0 on every claim-push in gateway.log."],"workaround":null,"fixed_in":null,"score":56.1,"first_seen":"2026-10-19T18:38:33Z"},{"id":"anomalyco/opencode#50885","harness":"opencode","harness_name":"OpenCode","repo":"anomalyco/opencode","number":50885,"title":"NO API KEY","url":"https://github.com/anomalyco/opencode/issues/50885","state":"closed","state_reason":"completed","labels":[],"created":"2026-09-23T10:46:44Z","updated":"2026-10-17T20:47:19Z","closed":"2026-10-01T03:51:35Z","reactions":11,"thumbs_up":11,"comments":10,"severity":"medium","severity_reasons":["engagement 21","class:auth/session"],"topic":"auth/session","classes":["auth/session"],"cosmetic_ui":false,"summary":"I subscribed to OpenCode Go today. My Go subscription is active, but the console does not show any personal API key or any option to create/copy one. /connect → OpenCode Go requires an API key. The Keys page only allows creating Service…","versions":["2.0.15"],"error_snippets":[],"workaround":null,"fixed_in":null,"score":56.1,"first_seen":"2026-10-19T18:38:33Z"},{"id":"anomalyco/opencode#51419","harness":"opencode","harness_name":"OpenCode","repo":"anomalyco/opencode","number":51419,"title":"Incorrect API key provided: sk","url":"https://github.com/anomalyco/opencode/issues/51419","state":"closed","state_reason":"not_planned","labels":[],"created":"2026-09-25T23:07:16Z","updated":"2026-09-26T01:43:51Z","closed":"2026-09-26T01:43:51Z","reactions":12,"thumbs_up":12,"comments":8,"severity":"medium","severity_reasons":["engagement 20","class:auth/session"],"topic":"auth/session","classes":["auth/session"],"cosmetic_ui":false,"summary":"OpenCode returns an OpenAI 401 when sending a prompt. The error is: `Incorrect API key provided: sk-svcac OpenCode’s bot is asking you to edit the issue body (not just comment). The numbered fields were left as No response. You have about…","versions":[],"error_snippets":["OpenCode returns an OpenAI 401 when sending a prompt. The error is:","OpenCode’s bot is asking you to edit the issue body (not just comment). The numbered fields were left as _No response_. You have about two hours.","I do not have the issue URL, so I cannot push the edit for you. On GitHub: Edit the issue, then replace _No response_ under these headings.","Run this first and paste the exact output into OpenCode version:","Select an OpenAI chat model (note the exact provider/model id in this field)."],"workaround":null,"fixed_in":null,"score":56.1,"first_seen":"2026-10-19T18:38:33Z"},{"id":"google-gemini/gemini-cli#29892","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":29892,"title":"Headless mode prints [object Object] instead of the API error (looks like #13553 again)","url":"https://github.com/google-gemini/gemini-cli/issues/29892","state":"open","state_reason":null,"labels":["area/non-interactive","status/bot-triaged","effort/small"],"created":"2026-10-23T11:08:03Z","updated":"2026-10-23T11:15:29Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"medium","severity_reasons":["class:self-host/headless"],"topic":"self-host/headless","classes":["self-host/headless","install/update"],"cosmetic_ui":false,"summary":"When an API call fails in headless mode, the last line I get is: The actual error (in my case a Vertex ModelNotFoundError: Publisher model ... was not found) doesn't make it to that line. This looks like the same thing as #13553, which was…","versions":["0.65.0"],"error_snippets":["An unexpected critical error occurred:[object Object]","ModelNotFoundError: Publisher model ... was not found","throw event.value.error;","StructuredError","String(error)"],"workaround":{"text":"`export GOOGLE_GENAI_USE_VERTEXAI=true GOOGLE_CLOUD_PROJECT= GOOGLE_CLOUD_LOCATION=us-central1` 2.","source":"issue body","url":"https://github.com/google-gemini/gemini-cli/issues/29892"},"fixed_in":null,"score":55.8,"first_seen":"2026-10-24T08:54:58Z"},{"id":"openclaw/openclaw#171446","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":171446,"title":"[Bug]: `openclaw update` loops in `validating — updater-runtime-retention`, ignores `--timeout`, and does heavy disk writes (2026.9.9 → 2026.10.3)","url":"https://github.com/openclaw/openclaw/issues/171446","state":"closed","state_reason":"completed","labels":["P2","clawsweeper:no-new-fix-pr","clawsweeper:needs-maintainer-review","issue-rating: 🦪 silver shellfish","maturity:stable","impact:ux-friction"],"created":"2026-10-13T23:06:58Z","updated":"2026-10-19T02:19:15Z","closed":"2026-10-17T15:27:37Z","reactions":2,"thumbs_up":1,"comments":7,"severity":"medium","severity_reasons":["class:install/update","class:network"],"topic":"install/update","classes":["install/update","network","always-on/daemon","crash/hang","mcp/tools"],"cosmetic_ui":false,"summary":"OpenClaw: 2026.9.9 (bcfc888) → target 2026.10.3 Install: npm global package (installKind: package, npm 11.14.1) Node: v24.21.0 Gateway: systemd user service, channel stable openclaw update --tag 2026.10.3 --yes --timeout 600 installs the…","versions":["2026.9.9","2026.10.3"],"error_snippets":["\"packageActivationError\": \"managed handoff lease database identity changed\",","Warning: plugins.entries.acpx: Plugin \"acpx\" settings cannot be checked until its data/settings upgrade finishes. ...","exit=137 (after SIGINT)","openclaw update --tag 2026.10.3 --yes --timeout 600","GatewayStateOwnerContentionError"],"workaround":null,"fixed_in":null,"score":55.7,"first_seen":"2026-10-18T04:33:05Z"},{"id":"omacom/omarchy#5097","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":5097,"title":"Microsoft Edge crashes under Wayland with tile rule on hover","url":"https://github.com/omacom/omarchy/issues/5097","state":"open","state_reason":null,"labels":["bug"],"created":"2026-03-22T18:53:07Z","updated":"2026-10-08T00:25:46Z","closed":null,"reactions":4,"thumbs_up":0,"comments":6,"severity":"high","severity_reasons":["title:breakage"],"topic":"crash/hang","classes":["crash/hang"],"cosmetic_ui":false,"summary":"Microsoft Edge segfaults when hovering over images on certain sites (e.g. omarchythemes.com theme detail pages). The crash is caused by the interaction between Edge's native Wayland popup/overlay handling and the tile on window rule in…","versions":["0.54.2","1.5"],"error_snippets":["--enable-features=TouchpadOverscrollHistoryNavigation","--load-extension=~/.local/share/omarchy/default/chromium/extensions/copy-url","microsoft-edge-stable-flags.conf","microsoft-edge-flags.conf","--disable-extensions"],"workaround":{"text":"Create `~/.config/microsoft-edge-stable-flags.conf`:","source":"issue body","url":"https://github.com/omacom/omarchy/issues/5097"},"fixed_in":null,"score":55.7,"first_seen":"2026-10-22T09:27:06Z"},{"id":"google-gemini/gemini-cli#29338","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":29338,"title":"Rootless podman sandbox hit EACCES because --userns=keep-id is not set","url":"https://github.com/google-gemini/gemini-cli/issues/29338","state":"open","state_reason":null,"labels":["priority/p2","area/core","status/bot-triaged","kind/bug","good first issue","effort/medium"],"created":"2026-09-15T09:16:22Z","updated":"2026-10-25T19:00:17Z","closed":null,"reactions":0,"thumbs_up":0,"comments":8,"severity":"low","severity_reasons":["low engagement"],"topic":"self-host/headless","classes":["self-host/headless","install/update","auth/session"],"cosmetic_ui":false,"summary":"I'm running Gemini CLI locally on arch linux with the podman sandbox enabled: GEMINISANDBOX=podman When i run npm start, the CLI gets to the point where it rebuilds one of the native dependencies inside the sandbox, but the rebuild fails…","versions":["0.59.0-nightly.20260825.g812f7a2bc"],"error_snippets":["being unable to remove the","I'm running Gemini CLI locally on arch linux with the podman sandbox enabled: `GEMINI_SANDBOX=podman`","Because the project is bind-mounted into the container, this can result in files being owned by a UID that the host user can't modify. In my case, that eventually shows up as `node-gyp` being unable to remove the `build`","If the sandbox is using rootless podman and UID/GID mapping is enabled, I would expect the CLI to handle the user namespace mapping automatically so that bind-mounted files remain owned by host user.","This `/about` is from a local npm start after the keep-id fix. The original `EACCES` was the same nightly (`0.59.0-nightly.20260825.g812f7a2bc`) with `GEMINI_SANDBOX=podman` and no `SANDBOX_FLAGS`"],"workaround":{"text":"This is closely related to `https://github.com/google-gemini/gemini-cli/issues/15376` where `--userns=keep-id` was suggested as a workaround. the workaround does fixes the underlying UID mapping problem, but requiring users to manually add a podman specific flag feels like something the CLI could…","source":"issue body","url":"https://github.com/google-gemini/gemini-cli/issues/29338"},"fixed_in":null,"score":55.6,"first_seen":"2026-10-19T18:38:33Z"},{"id":"omacom/omarchy#8701","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":8701,"title":"QT_QPA_PLATFORM=wayland;xcb causes SteamVR vrmonitor to abort on startup","url":"https://github.com/omacom/omarchy/issues/8701","state":"open","state_reason":null,"labels":["bug"],"created":"2026-08-28T01:14:16Z","updated":"2026-10-09T05:30:57Z","closed":null,"reactions":4,"thumbs_up":3,"comments":7,"severity":"low","severity_reasons":["low engagement"],"topic":"install/update","classes":["install/update","crash/hang","mcp/tools"],"cosmetic_ui":false,"summary":"On Omarchy 4.0.0-1, SteamVR's vrmonitor aborts during Qt GUI initialization because the session-wide environment sets: SteamVR's bundled Qt 5 installation provides only the xcb platform plugin. It treats wayland;xcb as one literal plugin…","versions":["4.0.0-1"],"error_snippets":["This application failed to start because it could not find or load the Qt platform plugin \"wayland;xcb\".","then exits with SIGABRT. The coredump stack reaches","QMessageLogger::fatal()","/usr/share/omarchy/default/hypr/envs.lua","with no new Qt platform-plugin error or new"],"workaround":null,"fixed_in":null,"score":55.6,"first_seen":"2026-10-22T09:27:06Z"},{"id":"kirodotdev/Kiro#11850","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":11850,"title":"CLI: refresh tokens for remote MCP servers are refreshed in memory but never persisted, forcing re-auth on every restart","url":"https://github.com/kirodotdev/Kiro/issues/11850","state":"open","state_reason":"reopened","labels":["mcp","cli","bug"],"created":"2026-10-08T07:25:47Z","updated":"2026-10-12T14:33:26Z","closed":null,"reactions":0,"thumbs_up":0,"comments":3,"severity":"medium","severity_reasons":["class:self-host/headless","class:auth/session"],"topic":"self-host/headless","classes":["self-host/headless","auth/session","mcp/tools"],"cosmetic_ui":false,"summary":"What happens Add a remote OAuth MCP server and sign in. ~/.aws/sso/cache/ .token.json and .registration.json are written. Keep the session running past the access token's lifetime (5 minutes in our case). Kiro refreshes successfully in the…","versions":["2.28.0","2.1"],"error_snippets":["FailedBecauseTokenMightBeExpired","2. Keep the session running past the access token's lifetime (5 minutes in our case). Kiro refreshes successfully in the background: the server sees a refresh grant before each call, with no 401 first."],"workaround":null,"fixed_in":null,"score":55.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"anomalyco/opencode#52227","harness":"opencode","harness_name":"OpenCode","repo":"anomalyco/opencode","number":52227,"title":"Go API key is not available in the new v2 Console","url":"https://github.com/anomalyco/opencode/issues/52227","state":"closed","state_reason":"completed","labels":[],"created":"2026-09-30T05:04:44Z","updated":"2026-10-21T21:42:12Z","closed":"2026-10-21T11:38:24Z","reactions":6,"thumbs_up":4,"comments":14,"severity":"medium","severity_reasons":["engagement 20","class:auth/session"],"topic":"auth/session","classes":["auth/session"],"cosmetic_ui":false,"summary":"I have an active OpenCode Go subscription, and Go models work normally in OpenCode CLI. However, I cannot find any way to create, view, or copy a Go API key in the new v2 Console. The current Go documentation still says users can copy an…","versions":["1.18.33"],"error_snippets":["However, I cannot find any way to create, view, or copy a **Go API key** in the new v2 Console.","The current Go documentation still says users can copy an API key and use Go with external coding agents such as Claude Code."],"workaround":null,"fixed_in":null,"score":55.3,"first_seen":"2026-10-19T18:38:33Z"},{"id":"github/copilot-cli#5134","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":5134,"title":"Windows: runaway MCP OAuth initialization opens repeated browser sessions; Maximum update depth exceeded","url":"https://github.com/github/copilot-cli/issues/5134","state":"open","state_reason":null,"labels":["area:mcp"],"created":"2026-10-13T18:20:28Z","updated":"2026-10-14T05:25:00Z","closed":null,"reactions":1,"thumbs_up":1,"comments":0,"severity":"medium","severity_reasons":["class:install/update","class:auth/session"],"topic":"install/update","classes":["install/update","auth/session","mcp/tools","self-host/headless"],"cosmetic_ui":false,"summary":"Running copilot interactively on Windows causes repeated OAuth browser sessions during initialization. Browser sessions keep opening until the user terminates Copilot. This prevents normal startup and authentication. Existing CLI logs…","versions":["1.0.60"],"error_snippets":["\"url\": \"https://mcp.expo.dev/mcp\",","2026-10-13T18:16:02.994Z [ERROR] Server supabase-staging requires authentication, initiating OAuth flow","2026-10-13T18:16:03.071Z [ERROR] Server supabase-staging requires authentication, initiating OAuth flow","2026-10-13T18:16:03.094Z [ERROR] Server supabase-staging requires authentication, initiating OAuth flow","2026-10-13T18:16:03.099Z [ERROR] Server supabase-staging requires authentication, initiating OAuth flow"],"workaround":null,"fixed_in":null,"score":55.2,"first_seen":"2026-10-22T09:27:06Z"},{"id":"kirodotdev/Kiro#11669","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":11669,"title":"--auth-method cli ignored in ACP v3 mode — causes infinite auth loop with hosts that don't implement   _kiro/auth/getAccessToken","url":"https://github.com/kirodotdev/Kiro/issues/11669","state":"open","state_reason":null,"labels":["os: windows","mcp","auth","pending-maintainer-response","cli"],"created":"2026-09-29T10:17:16Z","updated":"2026-10-12T08:22:01Z","closed":null,"reactions":1,"thumbs_up":1,"comments":4,"severity":"medium","severity_reasons":["class:auth/session","class:mcp/tools"],"topic":"auth/session","classes":["auth/session","mcp/tools"],"cosmetic_ui":false,"summary":"When running kiro-cli acp --agent-engine v3 --auth-method cli, the --auth-method cli flag is ignored. The agent hardcodes acp-callback authentication regardless of the flag, causing an infinite retry loop when the ACP host doesn't…","versions":["2.25.0"],"error_snippets":["\"args\": [\"acp\", \"--agent-engine\", \"v3\", \"--auth-method\", \"cli\", \"-v\"]","When running kiro-cli acp --agent-engine v3 --auth-method cli, the --auth-method cli flag is ignored. The agent hardcodes acp-callback authentication regardless of the flag, causing an infinite retry loop when the ACP ho","This breaks v3 ACP integration with hosts like JetBrains AI Assistant that don't implement Kiro's proprietary auth callback extension.","Use --agent-engine v2 instead, which respects CLI credentials. However, this loses access to MCP tool integration","1. --auth-method cli should be respected in ACP mode, using the existing CLI credentials from kiro-cli login"],"workaround":{"text":"Use --agent-engine v2 instead, which respects CLI credentials. However, this loses access to MCP tool integration which requires v3.","source":"issue body","url":"https://github.com/kirodotdev/Kiro/issues/11669"},"fixed_in":null,"score":55.0,"first_seen":"2026-10-22T09:27:06Z"},{"id":"omacom/omarchy#14136","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":14136,"title":"Quickshell exits with status 137 on wake from suspend, briefly flashing Hyprland lockdead screen","url":"https://github.com/omacom/omarchy/issues/14136","state":"open","state_reason":null,"labels":[],"created":"2026-10-03T12:07:56Z","updated":"2026-10-18T06:37:09Z","closed":null,"reactions":2,"thumbs_up":2,"comments":4,"severity":"medium","severity_reasons":["class:always-on/daemon"],"topic":"always-on/daemon","classes":["always-on/daemon"],"cosmetic_ui":false,"summary":"Apple MacBookAir5,2, Intel Core i5-3427U, Intel HD Graphics 4000, Omarchy 4.0.4-1, Hyprland 0.56.2, Linux 7.2.5-3-omarchy Upon opening the laptop lid and resuming from suspend (S3), the Omarchy lock screen briefly displayed Hyprland's…","versions":["4.0.4-1","0.56.2"],"error_snippets":["Oct 03 13:53:18 omarchy-shell[11045]: Omarchy shell exited with status 137; relaunching.","Upon opening the laptop lid and resuming from suspend (S3), the Omarchy lock screen briefly displayed Hyprland's emergency `lockdead.png` failsafe screen (*\"Oopsie daisy, it looks like you locked your screen but the lock","Reviewing journalctl logs revealed that the running Quickshell lock client (PID 1663) abruptly terminated with status **137** (`SIGKILL`) while processing rapid post-wake D-Bus property changes from NetworkManager access","Because the active `ext-session-lock-v1` client died while the compositor was locked, Hyprland engaged its failsafe lock screen (`/usr/share/hypr/lockdead.png`)."],"workaround":null,"fixed_in":null,"score":54.8,"first_seen":"2026-10-19T18:38:33Z"},{"id":"omacom/omarchy#14635","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":14635,"title":"Menu is black after Qt 6.12.0 / quickshell 0.3.2 upgrade (Color singleton sub-objects undefined)","url":"https://github.com/omacom/omarchy/issues/14635","state":"closed","state_reason":"completed","labels":[],"created":"2026-10-08T22:48:07Z","updated":"2026-10-18T07:42:54Z","closed":"2026-10-18T07:42:54Z","reactions":4,"thumbs_up":4,"comments":9,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update","always-on/daemon","mcp/tools"],"cosmetic_ui":false,"summary":"After upgrading to Qt 6.12.0, Super + Space opens the Omarchy menu as a fully black screen. The shell's Color singleton sub-objects (Color.menu, Color.tooltip, Color.bar, ...) evaluate to undefined, so every themed surface (menu, tooltips…","versions":["6.12.0","0.3.2","4.0.4-1","0.56.2"],"error_snippets":["@plugins/bar/Bar.qml[75:3]: Unable to assign [undefined] to QColor","@Ui/PanelToolTip.qml[22:-1]: TypeError: Cannot read property 'background' of undefined","plugins/menu/Menu.qml[87:-1]: TypeError: Cannot read property 'background' of undefined","plugins/menu/Menu.qml[91:-1]: TypeError: Cannot read property 'scrim' of undefined","@Commons/Border.qml[25:-1]: TypeError: Cannot read property 'menu.border-alpha' of undefined"],"workaround":null,"fixed_in":null,"score":54.8,"first_seen":"2026-10-19T18:38:33Z"},{"id":"omacom/omarchy#14438","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":14438,"title":"Idle screensaver/lock never fire when shell.json idle timeout differs from defaults (Quickshell IdleMonitor timeout bug)","url":"https://github.com/omacom/omarchy/issues/14438","state":"open","state_reason":null,"labels":[],"created":"2026-10-07T12:38:25Z","updated":"2026-10-18T12:02:40Z","closed":null,"reactions":1,"thumbs_up":1,"comments":3,"severity":"medium","severity_reasons":["class:network"],"topic":"network","classes":["network","always-on/daemon","crash/hang"],"cosmetic_ui":false,"summary":"Intel i7-12800H, Intel Iris Xe + NVIDIA RTX A1000 Laptop GPU, Omarchy 4.0.4-1, Quickshell 0.3.1-1, Hyprland 0.56.2 If idle.screensaver or idle.lock in ~/.config/omarchy/shell.json differs from the built-in defaults so that the first idle…","versions":["4.0.4-1","0.3.1-1","0.56.2","0.3.1"],"error_snippets":["IdleMonitor { timeout: to; onIsIdleChanged: console.log(\"CHANGED idle=\" + isIdle) }","IdleMonitor { timeout: 10; onIsIdleChanged: console.log(\"CONTROL idle=\" + isIdle) }","timeout: root.firstIdleTimeoutSeconds","quickshell -p /usr/share/omarchy/shell ipc call idle disable && quickshell -p /usr/share/omarchy/shell ipc call idle enable","enabled: false"],"workaround":{"text":"`quickshell -p /usr/share/omarchy/shell ipc call idle disable && quickshell -p /usr/share/omarchy/shell ipc call idle enable` after the shell has loaded its config (and after each `shell.json` idle change).","source":"issue body","url":"https://github.com/omacom/omarchy/issues/14438"},"fixed_in":null,"score":54.4,"first_seen":"2026-10-25T09:11:37Z"},{"id":"kirodotdev/Kiro#11625","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":11625,"title":"Task tools are down again. task_update / task_list / task_get are all returning \"Tool not available\"","url":"https://github.com/kirodotdev/Kiro/issues/11625","state":"open","state_reason":null,"labels":["os: windows","specs","pending-maintainer-response","theme:unexpected-error"],"created":"2026-09-26T22:20:25Z","updated":"2026-10-18T19:21:09Z","closed":null,"reactions":1,"thumbs_up":1,"comments":7,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update","network"],"cosmetic_ui":false,"summary":"I'm on the latest version as of 9/26/2026. I constantly deal with an issue where the taskupdate tools are not functioning, which causes my sessions to not be able to update the status of tasks in the tasks.md file of my specs. Task tools…","versions":["1.1.70","5.3","15.4"],"error_snippets":["15.4 is done and verified — but I can't flip the checkbox yet because task_update is down again (the same recurring transient blip; it's cleared on a reconnect before).","While executing tasks in a spec."],"workaround":null,"fixed_in":null,"score":54.3,"first_seen":"2026-10-19T18:38:33Z"},{"id":"NousResearch/hermes-agent#130277","harness":"hermes","harness_name":"Hermes Agent","repo":"NousResearch/hermes-agent","number":130277,"title":"[Bug]: Desktop can't connect to a remote gateway since #127201: loopback renderer origin fails /api/ws Origin check","url":"https://github.com/NousResearch/hermes-agent/issues/130277","state":"closed","state_reason":"completed","labels":["type/bug","comp/cli","comp/gateway","area/auth","P1","sweeper:risk-security-boundary","comp/desktop"],"created":"2026-10-01T10:15:31Z","updated":"2026-10-01T20:01:24Z","closed":"2026-10-01T20:01:24Z","reactions":2,"thumbs_up":1,"comments":6,"severity":"high","severity_reasons":["label:p1"],"topic":"always-on/daemon","classes":["always-on/daemon","self-host/headless","auth/session","install/update","network"],"cosmetic_ui":false,"summary":"Since #127201 (d428df2a65, \"fix(desktop): restore YouTube embeds in the packaged app\"), the packaged Desktop renderer is served from http://127.0.0.1:47891. Its WebSocket upgrades therefore carry Origin: http://127.0.0.1:47891. When…","versions":["3.11.16"],"error_snippets":["'null' -> None","[renderer error:main] Hermes couldn't start: Error: Could not connect to Hermes gateway (WebSocket error before open)","at Tt.connectFailure (http://127.0.0.1:47891/assets/i18n-rJpzlE2h.js:2:12713)","or the bound host is accepted. Every","Hermes couldn't start: Error: Could not connect to Hermes gateway (WebSocket error before open)"],"workaround":null,"fixed_in":null,"score":54.1,"first_seen":"2026-10-18T04:32:26Z"},{"id":"github/copilot-cli#5124","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":5124,"title":"`--agent <name>` selection is dropped ~1s after startup (race with sign-in / MCP reload)","url":"https://github.com/github/copilot-cli/issues/5124","state":"open","state_reason":null,"labels":["area:agents"],"created":"2026-10-12T15:17:03Z","updated":"2026-10-20T11:08:50Z","closed":null,"reactions":0,"thumbs_up":0,"comments":2,"severity":"medium","severity_reasons":["class:auth/session","class:mcp/tools"],"topic":"auth/session","classes":["auth/session","mcp/tools","always-on/daemon","self-host/headless"],"cosmetic_ui":false,"summary":"When starting the CLI with --agent , the custom agent is selected and then silently deselected ~1 s later, before any user input. The banner shows Selected custom agent: my-agent, then MCP Servers reloaded: 1 server connected, and the…","versions":["11.603","11.765","11.717","12.400"],"error_snippets":["14:59:12.400 subagent.deselected {}","events.jsonl","Failed to start MCP Servers: session not found","It looks like a race between the startup agent selection and the agent/MCP refresh after sign-in. `events.jsonl` of an affected session (no user input in between):","4. Agent is gone from the status bar; `events.jsonl` shows `subagent.deselected`."],"workaround":{"text":"Frequency (automated pty launches, no input, 15 s each): dropped 5/6 with a user-level agent (`--agent my-agent`), 4/6 with a plugin agent (`--agent myplugin:my-agent`). - When MCP startup failed (`Failed to start MCP Servers: session not found`), the agent was never dropped, which points to the…","source":"issue body","url":"https://github.com/github/copilot-cli/issues/5124"},"fixed_in":null,"score":53.9,"first_seen":"2026-10-21T09:24:21Z"},{"id":"github/copilot-cli#3682","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":3682,"title":"Support refreshing the BYOK provider credential without restarting the CLI","url":"https://github.com/github/copilot-cli/issues/3682","state":"closed","state_reason":"completed","labels":["area:authentication","area:models"],"created":"2026-06-04T21:57:27Z","updated":"2026-09-29T08:25:57Z","closed":"2026-09-24T10:11:21Z","reactions":9,"thumbs_up":9,"comments":4,"severity":"critical","severity_reasons":["title:data-loss/security"],"topic":"auth/session","classes":["auth/session","always-on/daemon","network"],"cosmetic_ui":false,"summary":"When using a BYOK provider with a short-lived bearer credential (e.g. an Entra ID / Azure AD OAuth access token, an AWS STS token, or any OIDC JWT), the Copilot CLI reads COPILOTPROVIDERAPIKEY (and any custom headers) once at process…","versions":[],"error_snippets":["When using a BYOK provider with a short-lived bearer credential (e.g. an Entra ID / Azure AD OAuth access token, an AWS STS token, or any OIDC JWT), the Copilot CLI reads COPILOT_PROVIDER_API_KEY (and any custom headers)","Entra access tokens live ~60–90 minutes. In a long interactive coding session the token expires mid-conversation, the upstream gateway (e.g. Azure API Management running validate-jwt) starts returning HTTP 401, and the C","This forces BYOK deployments that want true per-user identity (rather than a long-lived static API key) to run an external local reverse-proxy sidecar purely to re-mint and inject a fresh token on each request. That side","Provide a way for the CLI to obtain a fresh credential per request (or on 401), e.g. one of:","File-backed credential. COPILOT_PROVIDER_API_KEY_FILE=/path/to/token where the CLI re-reads the file per request (an external az/cron/sidecar keeps it fresh). Simpler but weaker than option 1."],"workaround":{"text":"export COPILOT_PROVIDER_TYPE=azure export COPILOT_PROVIDER_BASE_URL=https://my-apim.azure-api.us/openai export COPILOT_MODEL=gpt-5.1 ### CLI runs this to get a token, and re-runs it automatically before expiry / on 401: export COPILOT_PROVIDER_API_KEY_COMMAND='az account get-access-token --scope…","source":"issue body","url":"https://github.com/github/copilot-cli/issues/3682"},"fixed_in":null,"score":53.4,"first_seen":"2026-10-23T09:23:24Z"},{"id":"omacom/omarchy#4120","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":4120,"title":"Thinkpad stalling after entering Omarchy credentials and rebooting","url":"https://github.com/omacom/omarchy/issues/4120","state":"closed","state_reason":"completed","labels":["bug"],"created":"2026-01-07T23:33:47Z","updated":"2026-09-12T17:25:06Z","closed":"2026-01-09T16:18:47Z","reactions":5,"thumbs_up":5,"comments":12,"severity":"critical","severity_reasons":["title:data-loss/security"],"topic":"auth/session","classes":["auth/session","install/update","network"],"cosmetic_ui":false,"summary":"Thinkpad T480, Omarchy 3.3.0 I upgraded to 3.3.0 today, and after I enter the login credentials, it waits for 2 minutes showing the login window, then reboots. I pressed Ctrl-Alt-7 to see the terminal while it was waiting, here is the…","versions":["3.3.0"],"error_snippets":["I upgraded to 3.3.0 today, and after I enter the login credentials, it waits for 2 minutes showing the login window, then reboots."],"workaround":null,"fixed_in":null,"score":53.4,"first_seen":"2026-10-19T18:38:33Z"},{"id":"kirodotdev/Kiro#11486","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":11486,"title":"Sign-in hangs after upgrade to 1.1.14 on Linux — browser never opens, extension host stuck on startup lock (works on 1.0.437)","url":"https://github.com/kirodotdev/Kiro/issues/11486","state":"open","state_reason":null,"labels":[],"created":"2026-09-20T15:00:14Z","updated":"2026-09-20T20:04:37Z","closed":null,"reactions":0,"thumbs_up":0,"comments":4,"severity":"high","severity_reasons":["title:breakage"],"topic":"install/update","classes":["install/update","crash/hang","auth/session","mcp/tools","self-host/headless"],"cosmetic_ui":false,"summary":"Arch Linux - kernel version: 7.2.4-arch1-2 After upgrading kiro-ide from 1.0.437 to 1.1.14 on Arch Linux, sign-in is broken. Clicking \"Sign in\" shows \"We will take you to your default web browser to sign in and bring you back when it's…","versions":["1.1.14","1.0.437","7.2.4-arch1"],"error_snippets":["[11595:0920/072821.082365:INFO:CONSOLE:442] \"%cDEBUG background: #eee; color: #888 Comments: URIs of continue on comments to add to storage .\", source: vscode-file://vscode-app/opt/Kiro/resources/app/out/vs/workbench/wor","[main 2026-09-20T14:28:23.259Z] #2: https://prod.download.desktop.kiro.dev/stable/metadata-linux-x64-stable.json - begin undefined OI { original: {}, headers: [Object: null prototype] {} }","[main 2026-09-20T14:28:23.350Z] #2: https://prod.download.desktop.kiro.dev/stable/metadata-linux-x64-stable.json - end undefined 200 {","etag: '\"9d60d0ccc54dac509d0026458bfd68d6\"',","'x-amz-server-side-encryption': 'aws:kms',"],"workaround":{"text":"Downgrading to 1.0.437 resolves the issue.","source":"issue body","url":"https://github.com/kirodotdev/Kiro/issues/11486"},"fixed_in":null,"score":52.7,"first_seen":"2026-10-19T18:38:33Z"},{"id":"anomalyco/opencode#55250","harness":"opencode","harness_name":"OpenCode","repo":"anomalyco/opencode","number":55250,"title":"tui: false \"Session location unavailable\" prompt after network reconnect","url":"https://github.com/anomalyco/opencode/issues/55250","state":"closed","state_reason":"completed","labels":["reproduced"],"created":"2026-10-13T22:46:58Z","updated":"2026-10-21T02:09:43Z","closed":"2026-10-21T02:09:43Z","reactions":3,"thumbs_up":3,"comments":6,"severity":"medium","severity_reasons":["class:network"],"topic":"network","classes":["network","self-host/headless","install/update","crash/hang","mcp/tools"],"cosmetic_ui":false,"summary":"When the TUI reconnects after a brief network interruption, it can replace the composer with the \"Session location unavailable — Choose another directory to continue this session\" recovery prompt even though the session's directory is…","versions":["2.0.27"],"error_snippets":["onCleanup(client.event.on(\"server.connected\", () => sync(ref())))","currentLocation.error?.location.directory === session()!.location.directory","recovery prompt even though the session's directory is valid and unchanged. Selecting the","same directory resumes the session. No restart or upgrade is involved; any single failed","request in the post-reconnect location sync is enough to surface the false prompt."],"workaround":null,"fixed_in":null,"score":52.7,"first_seen":"2026-10-19T18:38:33Z"},{"id":"cline/cline#14091","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":14091,"title":"[CLI] Massive memory leak (5GB+ RAM, 11GB+ swap) and SSD-killing SQLite write churn during streaming with reasoning models","url":"https://github.com/cline/cline/issues/14091","state":"open","state_reason":null,"labels":["CLI"],"created":"2026-09-13T00:11:50Z","updated":"2026-10-21T18:36:41Z","closed":null,"reactions":1,"thumbs_up":1,"comments":4,"severity":"high","severity_reasons":["title:breakage"],"topic":"crash/hang","classes":["crash/hang","always-on/daemon","auth/session","network"],"cosmetic_ui":false,"summary":"I am using a beta version of Cline During a single prompt session using the Cline CLI with a reasoning model (z-ai/glm-5.3-flash), the Cline daemon experienced an extreme memory allocation explosion and disk write churn that completely…","versions":["3.0.61","5.3-flash","74.6","5.1"],"error_snippets":["~/.cline/data/db/hub-events-hub-production.db","hub-events-hub-production.db","monitor query failed with exit code 6) and the desktop shell (","During a single prompt session using the Cline CLI with a reasoning model (`z-ai/glm-5.3-flash`), the Cline daemon experienced an extreme memory allocation explosion and disk write churn that completely exhausted host RA","Reasoning models stream intermediate thinking/reasoning chunks (`reasoning.delta`). For every single token/delta received:"],"workaround":null,"fixed_in":null,"score":52.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"google-gemini/gemini-cli#29796","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":29796,"title":"Regression & Broken Ecosystem Workflow: Deprecating OAuth for Google AI Pro / Individual Accounts Leaves No Public Quota API for Developers","url":"https://github.com/google-gemini/gemini-cli/issues/29796","state":"open","state_reason":null,"labels":["status/need-triage","area/platform"],"created":"2026-10-17T10:06:17Z","updated":"2026-10-19T10:42:46Z","closed":null,"reactions":0,"thumbs_up":0,"comments":2,"severity":"high","severity_reasons":["title:breakage"],"topic":"install/update","classes":["install/update","auth/session"],"cosmetic_ui":false,"summary":"With the recent deprecation of Gemini CLI OAuth for individual, Google AI Pro, and Ultra accounts (\"Google no longer supports Gemini CLI OAuth for individual, AI Pro, or Ultra accounts. Enable CodexBar's Antigravity...\"), individual paying…","versions":[],"error_snippets":["\"Google no longer supports Gemini CLI OAuth for individual, AI Pro, or Ultra accounts. Enable CodexBar's Antigravity...\"","With the recent deprecation of Gemini CLI OAuth for individual, Google AI Pro, and Ultra accounts (`\"Google no longer supports Gemini CLI OAuth for individual, AI Pro, or Ultra accounts. Enable CodexBar's Antigravity...\"","While enterprise and Google Workspace accounts retain Gemini Code Assist OAuth access, paying consumer subscribers (Google One AI Premium / Pro / Ultra) are left with:","3. **Forced and friction-heavy migration**: Directing consumer developers to Antigravity without a unified, transparent quota endpoint or OAuth parity creates unnecessary ecosystem fragmentation.","1. **Provide a lightweight, read-only Quota / Usage endpoint for Google Accounts**: Even if interactive Code Assist features are restructured, allow individual and Google AI Pro users to authenticate via OAuth or an API "],"workaround":null,"fixed_in":null,"score":52.4,"first_seen":"2026-10-19T18:38:33Z"},{"id":"anomalyco/opencode#49903","harness":"opencode","harness_name":"OpenCode","repo":"anomalyco/opencode","number":49903,"title":"run: --command removed in v2 and slash commands are not expanded — no CLI path for headless command invocation","url":"https://github.com/anomalyco/opencode/issues/49903","state":"open","state_reason":null,"labels":[],"created":"2026-09-19T05:10:26Z","updated":"2026-09-19T05:10:26Z","closed":null,"reactions":5,"thumbs_up":5,"comments":0,"severity":"medium","severity_reasons":["class:self-host/headless"],"topic":"self-host/headless","classes":["self-host/headless"],"cosmetic_ui":false,"summary":"OpenCode 2.0.8 removed --command from opencode run, and run does not expand slash commands in the message, so there is no CLI way to invoke a command headlessly. --command was the supported headless path in v1 — see #7345, where the…","versions":["2.0.8"],"error_snippets":["... we expect the flag\". The migrate-v1 guide also says command definitions in","ERROR: Unrecognized flag: --command in command opencode run","OpenCode 2.0.8 removed `--command` from `opencode run`, and `run` does not expand slash commands in the message, so there is no CLI way to invoke a command headlessly.","`--command` was the supported headless path in v1 — see #7345, where the response was \"use the `--command mycmd` ... we expect the flag\". The migrate-v1 guide also says command definitions in `.opencode/` \"should continu","Observations from the same environment (2.0.8):"],"workaround":null,"fixed_in":null,"score":52.3,"first_seen":"2026-10-19T18:38:33Z"},{"id":"kirodotdev/Kiro#5060","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":5060,"title":"Environment variable substitution does not work for mcp.json","url":"https://github.com/kirodotdev/Kiro/issues/5060","state":"closed","state_reason":"not_planned","labels":["mcp","cli"],"created":"2026-01-21T10:19:38Z","updated":"2026-09-23T20:39:17Z","closed":"2026-09-23T20:39:16Z","reactions":6,"thumbs_up":6,"comments":9,"severity":"medium","severity_reasons":["engagement 15","class:mcp/tools"],"topic":"mcp/tools","classes":["mcp/tools","always-on/daemon","self-host/headless","auth/session"],"cosmetic_ui":false,"summary":"Ubuntu 24.04 (WSL) and MacOS 15.7.3 Error 'Mcp error: -32002: Client error: HTTP status client error (405 Method Not Allowed) for url (https://my-gw-rrin.gateway.bedrock-agentcore.us-west-1.amazonaws.com/mcp) Before running kiro-cli, env…","versions":["1.24.0"],"error_snippets":["export JWT_TOKEN=token_value","Error 'Mcp error: -32002: Client error: HTTP status client error (405 Method Not Allowed) for url (https://my-gw-rrin.gateway.bedrock-agentcore.us-west-1.amazonaws.com/mcp)","Before running `kiro-cli`, env var is defined with `export JWT_TOKEN=\"my_token\" `","if I put token inside .kiro/settings/mcp.json instead of `${JWT_TOKEN}` everything is working as expected.","Document which describes how it works https://kiro.dev/docs/cli/mcp/configuration/#environment-variables"],"workaround":null,"fixed_in":null,"score":52.2,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openclaw/openclaw#157760","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":157760,"title":"[Bug]: secrets audit flags a plaintext API key that configure --section model itself writes, with no supported non-Vault way to store a SecretRef in a SQLite auth profile","url":"https://github.com/openclaw/openclaw/issues/157760","state":"open","state_reason":null,"labels":["P2","clawsweeper:no-new-fix-pr","clawsweeper:needs-maintainer-review","clawsweeper:needs-product-decision","clawsweeper:needs-security-review","clawsweeper:source-repro","impact:security","impact:auth-provider","issue-rating: 🦞 diamond lobster"],"created":"2026-09-25T00:21:55Z","updated":"2026-10-06T02:54:33Z","closed":null,"reactions":2,"thumbs_up":1,"comments":4,"severity":"critical","severity_reasons":["label:security"],"topic":"auth/session","classes":["auth/session","always-on/daemon","install/update","crash/hang","mcp/tools"],"cosmetic_ui":false,"summary":"Behavior bug (incorrect output/state without crash) openclaw secrets audit reports PLAINTEXTFOUND for a provider API key held in a SQLite auth profile, but that plaintext is written by the official openclaw configure --section model flow…","versions":[],"error_snippets":["Config validation failed: <root>: Unrecognized key: \"auth-profiles:main:profiles\"","`openclaw secrets audit` reports `PLAINTEXT_FOUND` for a provider API key held in a SQLite auth profile, but that plaintext is written by the official `openclaw configure --section model` flow, and no shipped non-Vault c","1. Configure a provider credential with the documented flow: `openclaw configure --section model` (DeepSeek → enter API key). It reports `Updated config: ~/.openclaw/openclaw.json`.","4. Try to convert the entry to a ref with the generic SecretRef builder, targeting the auth-profile path form the docs describe:","The plaintext remains and cannot be removed through shipped tooling. The migration is also not sticky: any later run of the official credential flow rewrites the raw key, so the audit finding returns. The only documented"],"workaround":null,"fixed_in":null,"score":51.4,"first_seen":"2026-10-19T18:38:33Z"},{"id":"omacom/omarchy#14564","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":14564,"title":"Edge repo serves stale package database: hype missing, so omarchy update tries to replace it with an unrelated AUR package","url":"https://github.com/omacom/omarchy/issues/14564","state":"closed","state_reason":"completed","labels":["bug"],"created":"2026-10-08T09:32:27Z","updated":"2026-10-16T12:54:49Z","closed":"2026-10-12T06:18:03Z","reactions":4,"thumbs_up":4,"comments":9,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update","network"],"cosmetic_ui":false,"summary":"AMD Ryzen 7 6800U, Radeon 680M (integrated), Omarchy 4.0.0.r6807.g902fd8a-1 (edge channel; now rolled back to r6720.g8e02fc8-1) Since around 2026-10-08 09:04 UTC, the edge repo database at https://pkgs.omarchy.org/edge/x8664/omarchy.db…","versions":["0.4.2-1","0.4.3-1","6.12","2.0.4"],"error_snippets":["skill) while helping a user troubleshoot a failed","https://pkgs.omarchy.org/edge/x86_64/omarchy.db",", so they can't get fixes such as #14553 (Qt 6.12","[omarchy] Server = https://pkgs.omarchy.org/edge/$arch","installed from the edge repo and"],"workaround":null,"fixed_in":null,"score":51.1,"first_seen":"2026-10-19T18:38:33Z"},{"id":"github/copilot-cli#5185","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":5185,"title":"P3 - container-runtime/copilot-runtime-bazel-cache-memory-usage-vs-limits-dotcom on catalog_service:copilot-runtime-bazel-cache,kube_cluster:dotcom-3-gc-cus-02,kube_container_name:copilot-runtime-bazel-cache,kube_deployment:copilot-r... - 2026-10-18T14:11Z","url":"https://github.com/github/copilot-cli/issues/5185","state":"open","state_reason":null,"labels":["source:datadog","catalog_service:copilot-runtime-bazel-cache","CRITICAL"],"created":"2026-10-18T14:12:15Z","updated":"2026-10-22T07:04:32Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"critical","severity_reasons":["label:critical"],"topic":"always-on/daemon","classes":["always-on/daemon","self-host/headless","crash/hang"],"cosmetic_ui":false,"summary":"Source: [[P3] [Triggered] container-runtime/copilot-runtime-bazel-cache-memory-usage-vs-limits-dotcom on…","versions":["95.906"],"error_snippets":["Container `copilot-runtime-bazel-cache` in deployment `copilot-runtime-bazel-cache` (namespace `copilot-runtime-bazel-cache-production`, service `copilot-runtime-bazel-cache`) in cluster `dotcom-3-gc-cus-02` — its busies","3. If this is expected, right-size the memory request/limit; otherwise escalate to the service owner.","1. Pick ONE of the following exclusions:","[Monitor Status] · [Edit Monitor] · [Related Logs]"],"workaround":null,"fixed_in":null,"score":51.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"continuedev/continue#12902","harness":"continue","harness_name":"Continue","repo":"continuedev/continue","number":12902,"title":"UI Bug: Chat view container overflows and covers VS Code line numbers as chat history grows during resizing","url":"https://github.com/continuedev/continue/issues/12902","state":"closed","state_reason":"completed","labels":["area:chat","kind:bug","ide:vscode","os:windows"],"created":"2026-06-26T07:59:37Z","updated":"2026-10-10T15:29:22Z","closed":"2026-10-10T15:29:22Z","reactions":10,"thumbs_up":6,"comments":9,"severity":"medium","severity_reasons":["engagement 19","class:self-host/headless"],"topic":"self-host/headless","classes":["self-host/headless"],"cosmetic_ui":false,"summary":"I've tried finding an answer on the Continue docs site I'm not able to find an open issue that reports the same bug I've seen the troubleshooting guide on the Continue Docs When using the Continue extension in the primary/secondary side…","versions":["2.1.0"],"error_snippets":["When using the Continue extension in the primary/secondary side bar, resizing the container causes the Chat View to permanently \"eat\" a fixed width and bleed into the editor area, covering the line numbers/gutter.","4. Notice that the Chat view UI is now permanently oversized and overflows onto the editor's line numbers. The more content in the chat, the wider the \"eaten\" space becomes."],"workaround":null,"fixed_in":null,"score":50.9,"first_seen":"2026-10-19T18:38:33Z"},{"id":"google-gemini/gemini-cli#25283","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":25283,"title":"OAuth/Policy Error: \"Tool execution denied by policy\" prevents basic read/write tool usage","url":"https://github.com/google-gemini/gemini-cli/issues/25283","state":"closed","state_reason":"not_planned","labels":["priority/p1","area/security","Stale","status/bot-triaged","kind/bug","type/bug"],"created":"2026-04-13T13:36:05Z","updated":"2026-10-12T02:07:45Z","closed":"2026-10-12T02:07:45Z","reactions":9,"thumbs_up":9,"comments":8,"severity":"critical","severity_reasons":["label:security"],"topic":"auth/session","classes":["auth/session"],"cosmetic_ui":false,"summary":"What happened? The Gemini CLI is consistently throwing a Tool execution denied by policy error. This is a generalized blockade affecting all tool invocations, but it severely impacts fundamental I/O operations (file reading, file writing…","versions":["0.37.1"],"error_snippets":["Tool execution denied by policy","to enter the interactive CLI, then run the","The problem persists even after forcefully revoking and regenerating application default credentials or reinstalling gemini cli. The issue seems to stem from an overly restrictive \"fail-closed\" default architecture regar","The CLI should successfully execute read/write file operations and shell commands when requested by the prompt. If there is a missing scope or local configuration preventing tool usage, the CLI should output a clear, act","Run `gemini` to enter the interactive CLI, then run the `/about` command."],"workaround":null,"fixed_in":null,"score":50.6,"first_seen":"2026-10-19T18:38:33Z"},{"id":"kirodotdev/Kiro#11394","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":11394,"title":"OAuth-protected MCP servers unusable in `--no-interactive` mode: headless client lacks ACP secret-storage capability","url":"https://github.com/kirodotdev/Kiro/issues/11394","state":"open","state_reason":null,"labels":[],"created":"2026-09-15T09:24:00Z","updated":"2026-09-15T13:20:52Z","closed":null,"reactions":0,"thumbs_up":0,"comments":4,"severity":"high","severity_reasons":["title:breakage"],"topic":"self-host/headless","classes":["self-host/headless","auth/session","mcp/tools"],"cosmetic_ui":false,"summary":"macOS 26.7 (arm64) A remote MCP server that authenticates via OAuth connects reliably in interactive kiro-cli chat, but always fails with Unauthorized in kiro-cli chat --no-interactive — even when a valid, unexpired access token (plus a…","versions":["2.21.4"],"error_snippets":["{\"level\":\"error\",\"message\":\"Failed to connect\",\"serverName\":\"newrelic\",\"error\":\"Unauthorized\"}","A remote MCP server that authenticates via OAuth connects reliably in interactive `kiro-cli chat`, but **always** fails with `Unauthorized` in `kiro-cli chat --no-interactive` — even when a valid, unexpired access token ","Root cause appears to be that the non-interactive client does not advertise the ACP **secret storage** capability, so the stored MCP OAuth credentials are never read, and without **openExternalUrl** no OAuth flow can be ","`--no-interactive` sessions never read stored credentials and never attempt OAuth. Every connection attempt to the OAuth-protected server fails:"],"workaround":null,"fixed_in":null,"score":50.2,"first_seen":"2026-10-19T18:38:33Z"},{"id":"omacom/omarchy#14836","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":14836,"title":"Startup cursor restore never runs after a reload wipes its Lua state; pointer stays invisible","url":"https://github.com/omacom/omarchy/issues/14836","state":"open","state_reason":null,"labels":["bug"],"created":"2026-10-10T09:39:06Z","updated":"2026-10-22T08:29:01Z","closed":null,"reactions":2,"thumbs_up":2,"comments":2,"severity":"low","severity_reasons":["low engagement"],"topic":"other","classes":[],"cosmetic_ui":false,"summary":"Omarchy dev channel (dev-linked checkout, integration build db439093), Hyprland 0.56.2, AMD Ryzen AI 9 HX 370 w/ Radeon 890M (GPD G1628-04). Two displays configured: internal eDP-1 panel plus a Vestel TV on DP-1. Session uses a plain…","versions":["0.56.2"],"error_snippets":["hyprctl eval","ever runs — including the final one #14610 adds. The 15 s timer recovery is gone too:","registrations do not survive a reload either. At the time of probing, DP-1 was no longer in","Probed the compositor's Lua state directly (via `hyprctl eval` writing through `io.open`, same method as the reload probe upthread in #14512):"],"workaround":null,"fixed_in":null,"score":50.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"google-gemini/gemini-cli#29817","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":29817,"title":"GeminiCLI.com Feedback: madeinshow/mcp-gemini-extension missing from the Extensions Gallery (topic + root manifest + release, absent after the 2026-10-16 crawl)","url":"https://github.com/google-gemini/gemini-cli/issues/29817","state":"open","state_reason":null,"labels":["area/extensions","status/bot-triaged","effort/small"],"created":"2026-10-18T20:24:17Z","updated":"2026-10-19T10:03:36Z","closed":null,"reactions":0,"thumbs_up":0,"comments":2,"severity":"medium","severity_reasons":["class:mcp/tools"],"topic":"mcp/tools","classes":["mcp/tools","self-host/headless","install/update","auth/session"],"cosmetic_ui":false,"summary":"https://geminicli.com/extensions/ The made-in-show extension is not listed in the Extensions Gallery, nor present in https://geminicli.com/extensions.json, although the repository meets every documented publishing requirement and the…","versions":["0.1.2","0.1.1","0.59.0"],"error_snippets":["$ gemini extensions validate .","Extension . has been successfully validated.","$ gemini extensions install https://github.com/madeinshow/mcp-gemini-extension --consent --skip-settings","Extension \"made-in-show\" installed successfully and enabled.","$ gemini extensions list"],"workaround":null,"fixed_in":null,"score":49.8,"first_seen":"2026-10-19T18:38:33Z"},{"id":"cline/cline#15260","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":15260,"title":"Billing: accidental annual upgrade without confirmation — $82.48 refund requested","url":"https://github.com/cline/cline/issues/15260","state":"open","state_reason":null,"labels":["beta"],"created":"2026-10-22T02:08:29Z","updated":"2026-10-22T04:15:10Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update","auth/session"],"cosmetic_ui":false,"summary":"Cloud Platform N/A - ClinePass web billing issue I am using a beta version of Cline I am reporting a subscription billing issue and requesting a refund for an accidental annual upgrade. On October 21, 2026, I accidentally selected the…","versions":["82.48"],"error_snippets":["I requested a refund shortly after the charge, before using the annual subscription. I have already emailed support@cline.bot with my transaction details and followed up on Discord.","My refund email included the receipt and invoice details needed to identify the transaction. However, I have not yet received a response to my refund request through either email or Discord. Please confirm that my email ","Expected behavior:","Before charging for an annual subscription, the billing flow should clearly display the billing period and total charge, require explicit confirmation, and allow the user to cancel before payment.","Why did your system upgrade my monthly subscription to an annual plan and charge me $82.48 without a separate confirmation step asking me to explicitly confirm the annual billing period and the total amount to be charged"],"workaround":null,"fixed_in":null,"score":49.8,"first_seen":"2026-10-22T09:27:06Z"},{"id":"continuedev/continue#8826","harness":"continue","harness_name":"Continue","repo":"continuedev/continue","number":8826,"title":"bug remote ssh VS code","url":"https://github.com/continuedev/continue/issues/8826","state":"closed","state_reason":"not_planned","labels":["area:context-providers","kind:bug","ide:vscode","stale"],"created":"2025-11-21T04:45:20Z","updated":"2026-10-20T01:46:19Z","closed":"2026-10-20T01:46:19Z","reactions":6,"thumbs_up":6,"comments":12,"severity":"medium","severity_reasons":["engagement 18","class:self-host/headless"],"topic":"self-host/headless","classes":["self-host/headless"],"cosmetic_ui":false,"summary":"I've tried using the \"Ask AI\" feature on the Continue docs site to see if the docs have an answer I'm not able to find a related conversation on GitHub discussions that reports the same bug I'm not able to find an open issue that reports…","versions":["1.2.10-linux","1.94.2","1.2.10"],"error_snippets":["Continue Extension Version | 1.2.10 (Based on logs)"],"workaround":null,"fixed_in":null,"score":49.8,"first_seen":"2026-10-19T18:38:33Z"},{"id":"anthropics/claude-code#97226","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":97226,"title":"[BUG] Remote Control sessions not accessible from claude.ai/code without session-specific URL","url":"https://github.com/anthropics/claude-code/issues/97226","state":"open","state_reason":null,"labels":["bug","area:claude-code-web","platform:web"],"created":"2026-09-25T18:11:54Z","updated":"2026-10-01T14:51:18Z","closed":null,"reactions":4,"thumbs_up":4,"comments":1,"severity":"medium","severity_reasons":["class:self-host/headless"],"topic":"self-host/headless","classes":["self-host/headless"],"cosmetic_ui":false,"summary":"Title: Remote Control sessions not accessible from claude.ai/code without session-specific URL The claude.ai/code landing page and the Code button in claude.ai both redirect to GitHub onboarding with no way to skip. This makes it…","versions":["2.1.28"],"error_snippets":["No error. Stuck in cloud environment onboarding screen","The landing page shows only the GitHub onboarding flow. There is no session list, no \"skip\" option, and no path to Remote Control sessions from the browser entry points.","This is a big blocker for enterprise adoption where cloud environments are not setup."],"workaround":null,"fixed_in":null,"score":49.6,"first_seen":"2026-10-26T10:12:02Z"},{"id":"google-gemini/gemini-cli#29461","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":29461,"title":"Sandboxed Gemini CLI loses login and session information since v0.60.0","url":"https://github.com/google-gemini/gemini-cli/issues/29461","state":"open","state_reason":null,"labels":["priority/p1","area/platform","status/bot-triaged","kind/bug","status/need-information"],"created":"2026-09-23T09:01:50Z","updated":"2026-10-14T19:05:10Z","closed":null,"reactions":1,"thumbs_up":1,"comments":7,"severity":"high","severity_reasons":["label:p1"],"topic":"auth/session","classes":["auth/session"],"cosmetic_ui":false,"summary":"Since v0.60.0, running Gemini CLI with sandboxing asks for authentication every time. The OAuth tokens are not saved persistently. What is worse, all sessions are lost: it is not possible to resume previous sessions, and new sessions are…","versions":["0.60.0","0.60.0-8139a7f87fbd"],"error_snippets":["to enter the interactive CLI, then run the","Since v0.60.0, running Gemini CLI with sandboxing asks for authentication every time. The OAuth tokens are not saved persistently.","If this is the intedend behavior, what is now the recommended workflow to avoid losing all session information every time Gemini CLI is launched?","Login and, especially, session information to be preserved across invocations of a sandboxed Gemini CLI.","Run `gemini` to enter the interactive CLI, then run the `/about` command."],"workaround":null,"fixed_in":null,"score":49.6,"first_seen":"2026-10-19T18:38:33Z"},{"id":"google-gemini/gemini-cli#29595","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":29595,"title":"ACP: an MCP tool permission request doesn't say which server or tool asked, and a client-supplied MCP server can't be pre-authorised","url":"https://github.com/google-gemini/gemini-cli/issues/29595","state":"open","state_reason":null,"labels":["area/non-interactive","status/bot-triaged","effort/small"],"created":"2026-10-02T00:17:25Z","updated":"2026-10-10T17:17:40Z","closed":null,"reactions":0,"thumbs_up":0,"comments":3,"severity":"medium","severity_reasons":["class:self-host/headless","class:auth/session"],"topic":"self-host/headless","classes":["self-host/headless","auth/session","mcp/tools"],"cosmetic_ui":false,"summary":"An ACP client starts a session and supplies its own MCP server in session/new's mcpServers (ours serves a shared task board to each agent). Every first call of every tool on that server then stops on a session/requestpermission, once per…","versions":["0.62.0"],"error_snippets":["(ours serves a shared task board to each agent). Every first call of every tool on that server then stops on a","security.enablePermanentToolApproval","entry into an MCP server config with","The client can't safely answer for the user, because the request doesn't say who is asking.","So a client can't tell its own server from any other server that happens to share the name, such as one defined in a workspace's own `.gemini/settings.json`. Matching the title is not an identity check, and answering \"al"],"workaround":null,"fixed_in":null,"score":49.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"omacom/omarchy#7263","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":7263,"title":"Dual-boot install fails when the free space is on a different disk than Windows itself","url":"https://github.com/omacom/omarchy/issues/7263","state":"closed","state_reason":"completed","labels":["bug"],"created":"2026-08-17T11:54:40Z","updated":"2026-10-23T08:56:24Z","closed":"2026-10-23T08:56:24Z","reactions":7,"thumbs_up":7,"comments":6,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"AMD 5900x, RX9070XT, Omarchy 4.0.0 Omarchy: 4.0.0 ISO: omarchy-4.0.0.iso, SHA256 verified against the published checksum Boot media: SanDisk 64 GB USB written with balenaEtcher, Secure Boot disabled Machine: two NVMe drives — Samsung SSD…","versions":["4.0.0","465.8"],"error_snippets":["mount: /mnt/boot: fsconfig() failed: Can't find a SQUASHFS superblock on nvme0n1p3.","dmesg(1) may have more information after failed mount system call.","mounting the ESP failed (exit 32)"],"workaround":null,"fixed_in":null,"score":49.2,"first_seen":"2026-10-19T18:38:33Z"},{"id":"github/copilot-cli#5170","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":5170,"title":"P3 - container-runtime/copilot-runtime-bazel-cache-memory-usage-vs-limits-dotcom on catalog_service:copilot-runtime-bazel-cache,kube_cluster:dotcom-2-gc-cus-02,kube_container_name:copilot-runtime-bazel-cache,kube_deployment:copilot-r... - 2026-10-16T17:35Z","url":"https://github.com/github/copilot-cli/issues/5170","state":"open","state_reason":null,"labels":["source:datadog","catalog_service:copilot-runtime-bazel-cache","CRITICAL"],"created":"2026-10-16T17:36:15Z","updated":"2026-10-22T07:04:29Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"critical","severity_reasons":["label:critical"],"topic":"always-on/daemon","classes":["always-on/daemon","self-host/headless","crash/hang"],"cosmetic_ui":false,"summary":"Source: [[P3] [Triggered] container-runtime/copilot-runtime-bazel-cache-memory-usage-vs-limits-dotcom on…","versions":["95.955"],"error_snippets":["Container `copilot-runtime-bazel-cache` in deployment `copilot-runtime-bazel-cache` (namespace `copilot-runtime-bazel-cache-production`, service `copilot-runtime-bazel-cache`) in cluster `dotcom-2-gc-cus-02` — its busies","3. If this is expected, right-size the memory request/limit; otherwise escalate to the service owner.","1. Pick ONE of the following exclusions:","[Monitor Status] · [Edit Monitor] · [Related Logs]"],"workaround":null,"fixed_in":null,"score":49.0,"first_seen":"2026-10-22T09:27:06Z"},{"id":"github/copilot-cli#5169","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":5169,"title":"P3 - container-runtime/copilot-runtime-bazel-cache-memory-usage-vs-limits-dotcom on catalog_service:copilot-runtime-bazel-cache,kube_cluster:dotcom-1-gc-cus-02,kube_container_name:copilot-runtime-bazel-cache,kube_deployment:copilot-r... - 2026-10-16T17:32Z","url":"https://github.com/github/copilot-cli/issues/5169","state":"open","state_reason":null,"labels":["source:datadog","catalog_service:copilot-runtime-bazel-cache","CRITICAL"],"created":"2026-10-16T17:33:16Z","updated":"2026-10-22T07:04:23Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"critical","severity_reasons":["label:critical"],"topic":"always-on/daemon","classes":["always-on/daemon","self-host/headless","crash/hang"],"cosmetic_ui":false,"summary":"Source: [[P3] [Triggered] container-runtime/copilot-runtime-bazel-cache-memory-usage-vs-limits-dotcom on…","versions":["96.507"],"error_snippets":["Container `copilot-runtime-bazel-cache` in deployment `copilot-runtime-bazel-cache` (namespace `copilot-runtime-bazel-cache-production`, service `copilot-runtime-bazel-cache`) in cluster `dotcom-1-gc-cus-02` — its busies","3. If this is expected, right-size the memory request/limit; otherwise escalate to the service owner.","1. Pick ONE of the following exclusions:","[Monitor Status] · [Edit Monitor] · [Related Logs]"],"workaround":null,"fixed_in":null,"score":49.0,"first_seen":"2026-10-22T09:27:06Z"},{"id":"cline/cline#14782","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":14782,"title":"Vercel AI Gateway repeatedly resolves to Runware after a 429","url":"https://github.com/cline/cline/issues/14782","state":"open","state_reason":null,"labels":["VS Code"],"created":"2026-10-02T16:43:15Z","updated":"2026-10-02T17:07:18Z","closed":null,"reactions":1,"thumbs_up":1,"comments":2,"severity":"medium","severity_reasons":["class:always-on/daemon","class:network"],"topic":"always-on/daemon","classes":["always-on/daemon","network"],"cosmetic_ui":false,"summary":"VSCode Extension I am using a beta version of Cline I ran into a weird routing issue while using Cline with deepseek/deepseek-v4.1-flash. The request started failing with a 429: What caught my attention was the routing metadata. The…","versions":["4.1.22"],"error_snippets":["Rate limit exceeded for deepseek/deepseek-v4.1-flash:","VSCode Extension","The request started failing with a 429:","The first error contained:","At the same time, each error showed:"],"workaround":null,"fixed_in":null,"score":48.6,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openclaw/openclaw#180688","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":180688,"title":"Update failure: pnpm-staging-preflight (2026.10.3)","url":"https://github.com/openclaw/openclaw/issues/180688","state":"closed","state_reason":"completed","labels":["P2","issue-rating: 🦪 silver shellfish","impact:ux-friction"],"created":"2026-10-22T14:24:41Z","updated":"2026-10-23T07:39:03Z","closed":"2026-10-23T07:38:17Z","reactions":2,"thumbs_up":0,"comments":5,"severity":"high","severity_reasons":["title:breakage"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"This report was explicitly reviewed and confirmed in OpenClaw. OpenClaw version: 2026.10.3 Node version: 24.21.0 Update action: CLI command: openclaw update Update target: exact target unavailable; mode: pnpm Failed phase…","versions":["2026.10.3","24.21.0"],"error_snippets":["This report was explicitly reviewed and confirmed in OpenClaw."],"workaround":null,"fixed_in":null,"score":48.5,"first_seen":"2026-10-23T09:23:24Z"},{"id":"google-gemini/gemini-cli#29322","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":29322,"title":"[Bug/Agent Regression] Autonomous Skill Execution Failure, Excuse-Making, and Lack of Agentic Initiative in Gemini 3.8 Flash vs Claude 3.5 Sonnet","url":"https://github.com/google-gemini/gemini-cli/issues/29322","state":"open","state_reason":null,"labels":["status/need-triage","area/agent"],"created":"2026-09-14T02:54:58Z","updated":"2026-10-23T02:54:45Z","closed":null,"reactions":0,"thumbs_up":0,"comments":3,"severity":"high","severity_reasons":["title:breakage"],"topic":"install/update","classes":["install/update","mcp/tools","self-host/headless","auth/session"],"cosmetic_ui":false,"summary":"During an interactive video localization and YouTube packaging workflow, the agent (Gemini 3.8 Flash) exhibited complete breakdown of agentic initiative, active instruction non-compliance, and degenerative conversational behavior. Passive…","versions":["2.0","2.12.2","0.50.0"],"error_snippets":["During an interactive video localization and YouTube packaging workflow, the agent (Gemini 3.8 Flash) exhibited complete breakdown of agentic initiative, active instruction non-compliance, and degenerative conversational","The user entered /humanizer repeatedly (over 40 times) expecting the agent to adopt custom skill guidelines. The agent took zero action to find or read SKILL.md despite having full filesystem tool capabilities (find_by_n","When confronted, the agent entered recursive apology loops (\"Tôi xin lỗi\", \"Lỗi của tôi\") instead of taking operational action.","Once manually forced to read SKILL.md (which explicitly forbids labeled lists under Section 19 and colons under Section 8), the agent explicitly declared in chat that it had abolished all colons and labels, yet in the ve","4. Excuse-Making and Blaming:"],"workaround":null,"fixed_in":null,"score":48.3,"first_seen":"2026-10-19T18:38:33Z"},{"id":"omacom/omarchy#5986","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":5986,"title":"SDDM greeter defaults to us keymap instead of the one selected on install","url":"https://github.com/omacom/omarchy/issues/5986","state":"closed","state_reason":"completed","labels":["bug"],"created":"2026-05-27T19:52:51Z","updated":"2026-10-04T14:51:12Z","closed":"2026-07-18T19:43:12Z","reactions":6,"thumbs_up":6,"comments":6,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update","crash/hang","auth/session"],"cosmetic_ui":false,"summary":"Omarchy 3.8.2 (updated 2026-05-27) Hyprland 0.55.2 (was 0.54.3) hyprlock 0.9.5-4 (was 0.9.5-2) SDDM 0.21.0-7 (was 0.21.0-6) SDDM greeter rejects correct password after 3.8.x update — greeter falls back to us keymap because installer only…","versions":["3.8.2","0.55.2","0.54.3","0.9.5-4"],"error_snippets":["May 27 18:02:50 thinkpad sddm-helper[10526]: [PAM] authenticate: Authentication failure","May 27 18:02:50 thinkpad sddm[1223]: Authentication error: SDDM::Auth::ERROR_AUTHENTICATION \"Authentication failure\"","May 27 18:03:00 thinkpad sddm-helper[10545]: [PAM] authenticate: Authentication failure","May 27 18:03:08 thinkpad sddm-helper[10549]: [PAM] authenticate: Authentication failure","May 27 18:03:22 thinkpad sddm-helper[10620]: [PAM] authenticate: Authentication failure"],"workaround":{"text":"Added the following to `/usr/share/sddm/hyprland.conf`:","source":"issue body","url":"https://github.com/omacom/omarchy/issues/5986"},"fixed_in":null,"score":48.3,"first_seen":"2026-10-21T09:24:21Z"},{"id":"anthropics/claude-code#94229","harness":"claude-code","harness_name":"Claude Code","repo":"anthropics/claude-code","number":94229,"title":"Session stuck showing \"Can't reach your computer\" when resuming existing conversation, while new conversations connect fine","url":"https://github.com/anthropics/claude-code/issues/94229","state":"open","state_reason":null,"labels":["bug","platform:windows","area:desktop"],"created":"2026-09-14T08:16:45Z","updated":"2026-09-15T18:36:39Z","closed":null,"reactions":3,"thumbs_up":2,"comments":2,"severity":"medium","severity_reasons":["class:crash/hang"],"topic":"crash/hang","classes":["crash/hang","network"],"cosmetic_ui":false,"summary":"When resuming an existing conversation in the Claude Code desktop app, the session gets stuck with: \"Can't reach your computer — It may be asleep or offline. This session will reconnect when it's back.\" The message never clears while…","versions":["10.0.19045"],"error_snippets":["When resuming an existing conversation in the Claude Code desktop app, the session gets stuck with:","\"Can't reach your computer — It may be asleep or offline. This session will reconnect when it's back.\"","Expected: resuming an existing conversation should reconnect to the local machine the same way a new conversation does, without getting stuck in an unreachable state."],"workaround":null,"fixed_in":null,"score":48.1,"first_seen":"2026-10-26T10:12:02Z"},{"id":"github/copilot-cli#5066","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":5066,"title":"Assisted permissions regression","url":"https://github.com/github/copilot-cli/issues/5066","state":"open","state_reason":null,"labels":["area:permissions"],"created":"2026-10-06T18:01:33Z","updated":"2026-10-07T05:28:16Z","closed":null,"reactions":1,"thumbs_up":1,"comments":3,"severity":"high","severity_reasons":["title:breakage"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"Recently assisted permissions mode has been requiring my approval for too many commands. This is mostly just a feeling that something has changed so sorry for being vague, but for example this powershell command to find files in the…","versions":[],"error_snippets":["Get-ChildItem -Recurse -Depth 3 -Filter *.?sproj | Where-Object { $_.Name -match 'web|manage' } | Select-Object -Expand FullName; Get-ChildItem -Directory | Select-Object -Expand Name","Recently assisted permissions mode has been requiring my approval for too many commands. This is mostly just a feeling that something has changed so sorry for being vague, but for example this powershell command to find ","Along with other commands to build and run the project when I explicitly asked for the project to be built and ran"],"workaround":null,"fixed_in":null,"score":48.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"cline/cline#15214","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":15214,"title":"ClinePass: the gateway ends a DeepSeek V4.1 Flash stream with an in-band AI_InvalidResponseDataError chunk, and the CLI ends the whole task with no retry","url":"https://github.com/cline/cline/issues/15214","state":"open","state_reason":null,"labels":["CLI"],"created":"2026-10-20T13:24:57Z","updated":"2026-10-20T13:25:25Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"medium","severity_reasons":["class:always-on/daemon"],"topic":"always-on/daemon","classes":["always-on/daemon","self-host/headless","install/update","crash/hang","mcp/tools","network"],"cosmetic_ui":false,"summary":"CLI (cline from npm, 3.0.71), headless: cline --json -P cline-pass -m cline-pass/deepseek-v4.1-flash -t 7200 --retries 5 \" \" We run the Cline CLI unattended to build apps through an MCP server (long tasks of 20 to 200 requests, mostly…","versions":["14.5","1.9"],"error_snippets":["data: {\"id\":\"gen_01M5CESD7DAFQQW8223ZRZ12QX\",\"object\":\"chat.completion.chunk\",\"created\":1792493272,\"model\":\"vmc/deepseek-v4p1-flash-contributor-fallbacks\",\"error\":{\"code\":\"error\",\"message\":\"Stream error occurred\",\"name\":","Stream error occurred","AI_InvalidResponseDataError","The operation timed out.","input). On 2026-10-19 and 2026-10-20, six tasks with ClinePass + DeepSeek V4.1 Flash ended in the middle of the work:"],"workaround":null,"fixed_in":null,"score":48.0,"first_seen":"2026-10-21T09:24:21Z"},{"id":"cline/cline#15003","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":15003,"title":"Cline Pass is broken: deepseek-v4.1-flash emits zero tool calls, the model invents results, and support is ignoring paying customers","url":"https://github.com/cline/cline/issues/15003","state":"open","state_reason":null,"labels":[],"created":"2026-10-10T11:35:38Z","updated":"2026-10-10T11:37:58Z","closed":null,"reactions":1,"thumbs_up":1,"comments":1,"severity":"high","severity_reasons":["title:breakage"],"topic":"mcp/tools","classes":["mcp/tools","crash/hang","auth/session"],"cosmetic_ui":false,"summary":"I pay for Cline Pass. sub1ULOU3JvJ1E14BGMWjDRrINh. You shipped a broken product, and now you've stopped answering. Somewhere between 2026-10-02 19:17Z and 2026-10-03 05:35Z, deepseek-v4.1-flash was moved off the DeepSeek channel: Same…","versions":["80.8","80.0"],"error_snippets":["Real case, 2026-10-10, session on `provider: cline-pass` / `model: cline-pass/deepseek-v4.1-flash`. Turns 82–85: **zero `tool/call` events.** What was in the reasoning instead:","MATH-500 level 5, 100 items per arm, paired n=95:","3. If you're not going to do either: refund the period and cancel me.","Two support emails. Zero replies. So it's public now."],"workaround":null,"fixed_in":null,"score":47.9,"first_seen":"2026-10-19T18:38:33Z"},{"id":"kirodotdev/Kiro#11915","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":11915,"title":"[Incident report] Kiro CLI self-update broke the install and removed the launcher","url":"https://github.com/kirodotdev/Kiro/issues/11915","state":"open","state_reason":null,"labels":["os: windows","cli","bug"],"created":"2026-10-13T01:08:03Z","updated":"2026-10-14T18:35:42Z","closed":null,"reactions":0,"thumbs_up":0,"comments":3,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update","always-on/daemon"],"cosmetic_ui":false,"summary":"<h2 id=\"summary\" style=\"color: rgb(204, 204, 204); font-family: -apple-system,BlinkMacSystemFont,Segoe UI,Roboto,Helvetica Neue,Arial,sans-serif; font-size: 28.8px; font-style: normal; font-variant: normal; font-weight: 400…","versions":["2.30.0"],"error_snippets":["ran an MSI for 2.30.0 about five seconds after the first run exited. The second run was still using","<code style=\"background-color: rgba(27, 31, 35, 0.5); border-bottom-left-radius: 3px; border-bottom-right-radius: 3px; border-top-left-radius: 3px; border-top-right-radius: 3px; color: rgb(170, 170, 170); padding-bottom:","status 1603 (error 1316). The install was left without its launcher <code style=\"background-color: rgba(27, 31, 35, 0.5); border-bottom-left-radius: 3px; border-bottom-right-radius: 3px; border-top-left-radius: 3px; bord","rollback. The second run (which was already executing from <code style=\"background-color: rgba(27, 31, 35, 0.5); border-bottom-left-radius: 3px; border-bottom-right-radius: 3px; border-top-left-radius: 3px; border-top-ri","Time | Source / event | Detail"],"workaround":null,"fixed_in":null,"score":47.8,"first_seen":"2026-10-19T18:38:33Z"},{"id":"github/copilot-cli#4910","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":4910,"title":"Non-interactive MCP tool call hangs after progress notification until idle timeout","url":"https://github.com/github/copilot-cli/issues/4910","state":"open","state_reason":null,"labels":["triage"],"created":"2026-09-19T18:43:00Z","updated":"2026-09-20T03:50:39Z","closed":null,"reactions":0,"thumbs_up":0,"comments":3,"severity":"high","severity_reasons":["title:breakage"],"topic":"crash/hang","classes":["crash/hang","mcp/tools","network","always-on/daemon","self-host/headless"],"cosmetic_ui":false,"summary":"In non-interactive Copilot CLI, an already-discovered Azure MCP tool call can hang after delivering its initial progress notification. No tool result is delivered, and the MCP WebSocket closes five minutes later with idle timeout. The…","versions":["1.0.87-0","3.0.0-beta.45"],"error_snippets":["18:27:42.559Z [DEBUG] [rust:mcp::client] MCP request enqueued {\"request_id\":\"Number(2)\"}","18:32:41.840Z [DEBUG] [rust:log] Received close frame ... reason: \"idle timeout\"","idle timeout","In non-interactive Copilot CLI, an already-discovered Azure MCP tool call can hang after delivering its initial progress notification. No tool result is delivered, and the MCP WebSocket closes five minutes later with `id","The identical tool and payload succeed immediately when invoked directly through another MCP client (VS Code), which suggests the failure is in the Copilot CLI MCP request/session path rather than the Azure best-practice"],"workaround":null,"fixed_in":null,"score":47.7,"first_seen":"2026-10-19T18:38:33Z"},{"id":"github/copilot-cli#1825","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":1825,"title":"Empty Input Schema Breaks Copilot CLI","url":"https://github.com/github/copilot-cli/issues/1825","state":"closed","state_reason":"completed","labels":["area:mcp","area:tools"],"created":"2026-03-04T11:30:19Z","updated":"2026-10-02T08:50:11Z","closed":"2026-09-29T10:32:53Z","reactions":10,"thumbs_up":10,"comments":3,"severity":"medium","severity_reasons":["class:mcp/tools"],"topic":"mcp/tools","classes":["mcp/tools"],"cosmetic_ui":false,"summary":"Tools with no input parameters use an empty JSON Schema for their Input Schema. This causes Copilot CLI to reject the tool entirely and fail on any prompt when the MCP server is connected. 0.0.365 Commit: 76d0881 To have an MCP tool with…","versions":["0.0.365"],"error_snippets":["✘ Model call failed: {\"message\":\"Invalid schema for function 'ai-game-developer-assets-prefab-save':","field. An empty object schema (with only","Tools with no input parameters use an empty JSON Schema for their Input Schema. This causes Copilot CLI to reject the tool entirely and fail on **any** prompt when the MCP server is connected.","Copilot is stuck producing this error again and again. Nothing else happens, doesn't matter what prompt to type. Only adding an input argument into the input json schema of the tool resolves the problem.","Copilot doesn't have the error about MCP tool with no arguments and can continue to operate and use it."],"workaround":null,"fixed_in":null,"score":47.6,"first_seen":"2026-10-19T18:38:33Z"},{"id":"cline/cline#14775","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":14775,"title":"Cannot open the desktop application after updating to 0.0.42 because \"Tauri invoke failed for get desktop_backend_endpoint: desktop backend endpoint not ready\"","url":"https://github.com/cline/cline/issues/14775","state":"open","state_reason":null,"labels":[],"created":"2026-10-02T11:58:12Z","updated":"2026-10-02T14:26:25Z","closed":null,"reactions":0,"thumbs_up":0,"comments":6,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"I am using a beta version of Cline I updated the latest version 0.0.42 of cline from 0.0.37, but I can't open my app because it always Tauri invoke failed for get desktopbackendendpoint: desktop backend endpoint not ready and I can't open…","versions":["0.0.42","0.0.37"],"error_snippets":["Tauri invoke failed for get desktop_backend_endpoint: desktop backend endpoint not ready","I updated the latest version 0.0.42 of cline from 0.0.37, but I can't open my app because it always `Tauri invoke failed for get desktop_backend_endpoint: desktop backend endpoint not ready` and I can't open my conversat"],"workaround":null,"fixed_in":null,"score":47.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openclaw/openclaw#169662","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":169662,"title":"[Bug]: macOS managed Gateway update rejects protected Bun --no-install command in installedServiceCLI","url":"https://github.com/openclaw/openclaw/issues/169662","state":"closed","state_reason":"completed","labels":["bug","no-stale","bug:behavior","clawsweeper:fix-shape-clear","clawsweeper:queueable-fix","clawsweeper:source-repro","P0","issue-rating: 🦞 diamond lobster","impact:ux-release-blocker"],"created":"2026-10-12T06:14:25Z","updated":"2026-10-12T19:11:27Z","closed":"2026-10-12T19:11:27Z","reactions":2,"thumbs_up":1,"comments":4,"severity":"critical","severity_reasons":["label:p0"],"topic":"always-on/daemon","classes":["always-on/daemon","install/update","crash/hang","auth/session","mcp/tools","network"],"cosmetic_ui":false,"summary":"Behavior bug (incorrect output/state without crash) The stock signed macOS app cannot inspect a managed Bun Gateway command containing --no-install because installedServiceCLI only skips memory flags; the native update/recovery screen…","versions":["2026.10.1"],"error_snippets":["The retained Gateway was not eligible for this runtime switch.","Gateway health: ok=true; plugins.errors=[]; plugins.unavailable=[]; modelRuntime.degraded=false.","No raw config, environment files, authentication material, usernames, private project names, or screenshots are included.","@@ -495,6 +495,9 @@","let entry = prefix.dropFirst().drop(while: {"],"workaround":null,"fixed_in":null,"score":47.4,"first_seen":"2026-10-18T04:33:05Z"},{"id":"github/copilot-cli#1680","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":1680,"title":"pwsh.exe hardcoded in 6 places - CLI completely unusable on Windows 11   with only PowerShell 5.1 (closed #411 is still broken in v0.0.417)","url":"https://github.com/github/copilot-cli/issues/1680","state":"closed","state_reason":"completed","labels":["area:platform-windows","area:tools"],"created":"2026-02-25T13:31:32Z","updated":"2026-09-24T10:09:40Z","closed":"2026-09-24T10:09:40Z","reactions":13,"thumbs_up":12,"comments":9,"severity":"high","severity_reasons":["title:breakage"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"Issue #411 was closed as \"notplanned\" in Oct 2025, but the bug is still present and getting worse — it now blocks the CLI from running any shell command at all on Windows machines that ship only with Windows PowerShell 5.1…","versions":["5.1","0.0.417","0.0.415","0.0.414"],"error_snippets":["ensureShellIsSupported","Issue #411 was closed as \"not_planned\" in Oct 2025, but the bug is still present and getting worse — it now blocks the CLI from running **any shell command at all** on Windows machines that ship only with Windows PowerSh","The error on every single command:","<exited with error: PowerShell 6+ (pwsh) is not available. Please install it from https://aka.ms/powershell (https://aka.ms/powershell). Error: Error: Command failed: pwsh.exe --version 'pwsh.exe' is not recognized as an","0.0.417 (also reproducible on 0.0.415, 0.0.414 — any recent version)"],"workaround":null,"fixed_in":null,"score":47.4,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aaif-goose/goose#13002","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":13002,"title":"GDK: exhausted empty-response retries restart indefinitely","url":"https://github.com/aaif-goose/goose/issues/13002","state":"open","state_reason":null,"labels":[],"created":"2026-10-19T12:27:18Z","updated":"2026-10-20T11:57:03Z","closed":null,"reactions":0,"thumbs_up":0,"comments":2,"severity":"low","severity_reasons":["low engagement"],"topic":"auth/session","classes":["auth/session"],"cosmetic_ui":false,"summary":"Describe the bug A GDK state machine using InferenceRunner can keep calling the provider indefinitely when the provider returns empty responses. The runner makes an initial attempt and three retries, then stores a hidden fallback message…","versions":[],"error_snippets":["operation::{Emitter, MachineEffect, RunStatus},","errors::ProviderError,","enum Effect {","impl From<Message> for Effect {","impl MachineEffect for Effect {"],"workaround":null,"fixed_in":null,"score":47.1,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aaif-goose/goose#12978","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":12978,"title":"Support OAuth 2.0 / OIDC JWT authentication across model providers (LiteLLM, OpenAI, Anthropic, etc.)","url":"https://github.com/aaif-goose/goose/issues/12978","state":"open","state_reason":null,"labels":[],"created":"2026-10-17T19:08:38Z","updated":"2026-10-21T19:32:57Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"medium","severity_reasons":["class:auth/session"],"topic":"auth/session","classes":["auth/session","always-on/daemon","self-host/headless"],"cosmetic_ui":false,"summary":"What problem would this solve? In many enterprise and team environments, access to LLM providers—whether through unified gateways like LiteLLM Proxy, corporate API gateways (Kong, Envoy, Apigee), or direct enterprise endpoints for OpenAI…","versions":["2.0"],"error_snippets":["401 Unauthorized","to opt into or configure this shared auth provider alongside existing static API key and","3. **Direct JWT / External Token Source Compatibility**:"],"workaround":null,"fixed_in":null,"score":47.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openclaw/openclaw#162554","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":162554,"title":"WhatsApp: PENDING_DELIVERY_NOTICE silently replaces the real reply instead of supplementing it, recurring far beyond restart boundaries","url":"https://github.com/openclaw/openclaw/issues/162554","state":"open","state_reason":null,"labels":["P1","clawsweeper:no-new-fix-pr","clawsweeper:needs-maintainer-review","impact:message-loss","issue-rating: 🦪 silver shellfish"],"created":"2026-10-01T08:38:43Z","updated":"2026-10-11T11:31:33Z","closed":null,"reactions":3,"thumbs_up":1,"comments":5,"severity":"high","severity_reasons":["label:p1"],"topic":"always-on/daemon","classes":["always-on/daemon"],"cosmetic_ui":false,"summary":"PENDINGDELIVERYNOTICE (\"I couldn't confirm whether my previous reply reached this chat...\") is being sent in place of the real agent reply, repeatedly, across many separate turns over 24+ hours — not just once after a restart as the…","versions":["2026.9.7","57.805+05","57.849+05","58.431+05"],"error_snippets":["2026-10-01T12:35:58.431+05:30 [whatsapp] Sent message 3EB022D8EB4E0B3C3DC8F6 -> sha256:5c6d7b7db189 (583ms)","const owedNotice = settled === \"unknown\" && (current === \"queued\" || current === \"unknown\") && pending.context && pending.intentId && existingNotice?.intentId !== pending.intentId && (!existingNotice || existingNotice.cr","dist/execution-DF3a_BT9.mjs","connectTimeoutMs","defaultQueryTimeoutMs"],"workaround":null,"fixed_in":null,"score":46.9,"first_seen":"2026-10-18T04:30:03Z"},{"id":"anomalyco/opencode#48628","harness":"opencode","harness_name":"OpenCode","repo":"anomalyco/opencode","number":48628,"title":"terminal does not clipboard read (kitty OSC 5522) for image paste over SSH","url":"https://github.com/anomalyco/opencode/issues/48628","state":"open","state_reason":null,"labels":["2.0"],"created":"2026-09-12T07:26:33Z","updated":"2026-09-15T06:20:32Z","closed":null,"reactions":3,"thumbs_up":3,"comments":2,"severity":"medium","severity_reasons":["class:self-host/headless"],"topic":"self-host/headless","classes":["self-host/headless"],"cosmetic_ui":false,"summary":"I use the TUI over SSH and can't paste images. Unsure if it's a bug or a gap in the design: the clipboard I want to read is on my laptop, but the TUI is running on the remote box. Add a read path to the terminal clipboard backend using…","versions":["0.0.0-beta"],"error_snippets":["I use the TUI over SSH and can't paste images. Unsure if it's a bug or a gap in the design: the clipboard I want to read is on my laptop, but the TUI is running on the remote box.","Add a read path to the terminal clipboard backend using kitty's clipboard protocol (OSC 5522). It's MIME-typed and travels in-band, so it works through SSH. It could slot in behind the existing `ClipboardService.read({ p"],"workaround":null,"fixed_in":null,"score":46.8,"first_seen":"2026-10-19T18:38:33Z"},{"id":"Aider-AI/aider#4340","harness":"aider","harness_name":"Aider","repo":"Aider-AI/aider","number":4340,"title":"Can't install on Ubntu 25.04 with Python 3.13 and pipx","url":"https://github.com/Aider-AI/aider/issues/4340","state":"open","state_reason":null,"labels":[],"created":"2025-07-12T11:52:10Z","updated":"2026-10-14T13:00:24Z","closed":null,"reactions":1,"thumbs_up":1,"comments":6,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"Since upgrading Ubuntu from 24.10 to 25.04 (and thus, from Python 3.12 to 3.13.3) , pipx install aider-chat fails with: /home/user/.local/state/pipx/log/cmd2025-07-1213.40.46piperrors.log contains this: I tried importing…","versions":["25.04","24.10","3.13.3","10.9"],"error_snippets":["Fatal error from pip prevented installation. Full pip output in file:","/home/user/.local/state/pipx/log/cmd_2025-07-12_13.40.46_pip_errors.log","Error installing aider-chat.","ERROR: Exception:","Traceback (most recent call last):"],"workaround":{"text":"As a workaround, you can try installing `aider-chat` using an earlier Python version managed by `pyenv`:","source":"comment by @adr-39 (4 reactions)","url":"https://github.com/Aider-AI/aider/issues/4340#issuecomment-3268631102"},"fixed_in":null,"score":46.8,"first_seen":"2026-10-19T18:38:33Z"},{"id":"openclaw/openclaw#179480","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":179480,"title":"[Bug]: SQLite store worker OOMs at startup — cron run maintenance loads all runs into memory before pruning (512 MB hardcoded cap)","url":"https://github.com/openclaw/openclaw/issues/179480","state":"closed","state_reason":"completed","labels":["impact:crash-loop","P0"],"created":"2026-10-21T12:41:04Z","updated":"2026-10-21T14:00:29Z","closed":"2026-10-21T13:04:20Z","reactions":2,"thumbs_up":1,"comments":1,"severity":"critical","severity_reasons":["label:p0"],"topic":"always-on/daemon","classes":["always-on/daemon","crash/hang","install/update"],"cosmetic_ui":false,"summary":"After upgrading, the gateway crash-looped (5+ boots). The SQLite store worker hit its V8 heap cap during startup maintenance and took the gateway down each time. Observed on a long-running install; not yet reproduced on a clean install…","versions":["2026.10.5","1.18"],"error_snippets":["maxOldGenerationSizeMb: 512","After upgrading, the gateway crash-looped (5+ boots). The SQLite store worker hit its V8 heap cap during startup maintenance and took the gateway down each time.","2. Create several cron jobs on a 2–3 min interval whose command always fails (e.g. a script that exits 1).","3. Let them run until each has ~2,000 failed runs in history. Our case: several jobs at ~2,005 failures each; the agent DB pre-startup-migration backup was ~1.18 GB.","Patched `maxOldGenerationSizeMb` 512 → 2048 in the dist file. The gateway booted with 0 OOMs since. The patch is lost on every update."],"workaround":null,"fixed_in":null,"score":46.7,"first_seen":"2026-10-22T09:27:06Z"},{"id":"google-gemini/gemini-cli#29325","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":29325,"title":"CLI 1.2.2: -p (print mode) hangs indefinitely on all inference - no error, no response, no timeout","url":"https://github.com/google-gemini/gemini-cli/issues/29325","state":"open","state_reason":null,"labels":["priority/p1","area/agent","kind/bug","status/manual-triage","effort/medium"],"created":"2026-09-14T07:35:43Z","updated":"2026-09-16T13:03:25Z","closed":null,"reactions":0,"thumbs_up":0,"comments":4,"severity":"high","severity_reasons":["label:p1","title:breakage"],"topic":"crash/hang","classes":["crash/hang","network","always-on/daemon","self-host/headless","install/update","auth/session","mcp/tools"],"cosmetic_ui":false,"summary":"agy -p \"Hi\" hangs indefinitely (5+ minutes, no response, no error, no timeout message) on all inference calls in print mode. The CLI starts, accepts the prompt, then never returns. This happens on the latest version (1.2.2) and was…","versions":["1.2.2"],"error_snippets":["error: interrupted","--print-timeout","`agy -p \"Hi\"` hangs indefinitely (5+ minutes, no response, no error, no timeout message) on **all inference calls** in print mode. The CLI starts, accepts the prompt, then never returns. This happens on the latest versio","3. Wait — it hangs forever with no output, no error, no timeout","The CLI should either:"],"workaround":null,"fixed_in":null,"score":46.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"github/copilot-cli#4931","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":4931,"title":"Cloud agent: the built-in Playwright MCP session is already closed at the first tool call, and recovery reports it as browser OAuth required","url":"https://github.com/github/copilot-cli/issues/4931","state":"open","state_reason":null,"labels":["triage"],"created":"2026-09-22T08:58:55Z","updated":"2026-10-13T21:23:22Z","closed":null,"reactions":1,"thumbs_up":1,"comments":2,"severity":"medium","severity_reasons":["class:auth/session","class:mcp/tools"],"topic":"auth/session","classes":["auth/session","mcp/tools","always-on/daemon","self-host/headless","network"],"cosmetic_ui":false,"summary":"Split out of #4378, which carries two faults. The registry policy 401 in that issue has a cause and a workaround. This one does not, and it survives that workaround. In a Copilot cloud agent session on a GHEC data residency tenant (…","versions":["0.0.40","1.0.0"],"error_snippets":["MCP server 'playwright': Tool call failed: Error: MCP request failed: Transport closed","MCP server 'playwright': Tool call failed after session recovery:","MCPOAuthBrowserRequiredError: Browser-based OAuth required for http://localhost:3100/mcp","[ERROR] Using deferred connection for default Playwright server","tools/list after 420 seconds HTTP/1.1 200 OK (21 tools)"],"workaround":null,"fixed_in":null,"score":46.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"kirodotdev/Kiro#11370","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":11370,"title":"SSH_AUTH_SOCK not inherited by shell tool in KiroCrew dashboard v0.6.0","url":"https://github.com/kirodotdev/Kiro/issues/11370","state":"open","state_reason":null,"labels":["terminal","ssh","cli"],"created":"2026-09-14T00:52:42Z","updated":"2026-09-14T21:25:03Z","closed":null,"reactions":0,"thumbs_up":0,"comments":4,"severity":"medium","severity_reasons":["class:self-host/headless","class:auth/session"],"topic":"self-host/headless","classes":["self-host/headless","auth/session","always-on/daemon","install/update","network"],"cosmetic_ui":false,"summary":"macOS 26.6.2 Kiro CLI 2.21.4 (first noticed after updating today) KiroCrew v0.6.0 (also updated today) The shell tool available to agents in the KiroCrew dashboard does not inherit SSHAUTHSOCK from the parent process environment. This…","versions":["0.6.0","2.21.4"],"error_snippets":["echo \"SSH_AUTH_SOCK=$SSH_AUTH_SOCK\"","The shell tool available to agents in the KiroCrew dashboard does not inherit SSH_AUTH_SOCK from the parent process environment. This breaks git push over SSH.","Evidence that this is KiroCrew-specific, not kiro-cli:","Both return empty/error despite the variable being correctly set in the KiroCrew terminal panel opened from the same app.","Expected: Shell tool subprocess inherits environment from the Electron app process, including SSH_AUTH_SOCK."],"workaround":null,"fixed_in":null,"score":46.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"kirodotdev/Kiro#11824","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":11824,"title":"kiro-cli V3 TUI is not usable over SSH + tmux; legacy V1 UI is the only workaround","url":"https://github.com/kirodotdev/Kiro/issues/11824","state":"open","state_reason":null,"labels":["cli"],"created":"2026-10-07T05:26:11Z","updated":"2026-10-07T17:30:13Z","closed":null,"reactions":0,"thumbs_up":0,"comments":4,"severity":"medium","severity_reasons":["class:self-host/headless"],"topic":"self-host/headless","classes":["self-host/headless","crash/hang"],"cosmetic_ui":false,"summary":"Amazon Linux 2023.12.20260930 (x8664), kernel 6.12.110-135.202.amzn2023.x8664. Accessed over SSH, running inside tmux (TERM=screen-256color). kiro-cli 2.28.0 The V3 TUI is not usable for me in a plain remote terminal. I run kiro-cli over…","versions":["2.28.0"],"error_snippets":["correctly in the same environment.","individual symptoms are already tracked separately and each has been triaged in","remote development setup, and the escape hatch is to opt out of it entirely.","4. Work normally: scroll back through earlier output, resize the pane or the local","5. Observe the scrollback, selection, and redraw failures described in the issues"],"workaround":null,"fixed_in":null,"score":46.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aaif-goose/goose#11582","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":11582,"title":"OAuth authorization URL is incorrect when using Keycloak with pre-registered OAuth client","url":"https://github.com/aaif-goose/goose/issues/11582","state":"open","state_reason":null,"labels":[],"created":"2026-08-26T08:36:32Z","updated":"2026-10-14T19:58:26Z","closed":null,"reactions":0,"thumbs_up":0,"comments":6,"severity":"medium","severity_reasons":["class:auth/session"],"topic":"auth/session","classes":["auth/session","self-host/headless","mcp/tools"],"cosmetic_ui":false,"summary":"Describe the bug I'm testing the pre-registered OAuth client support for streamablehttp extensions introduced in Goose v1.47.0. PR #11182: https://github.com/aaif-goose/goose/pull/11182 Commit: ea6dbf30b8e53ac5a623320eab1b77f61b604166 My…","versions":["1.47.0"],"error_snippets":["enabled: true","I'm testing the pre-registered OAuth client support for `streamable_http` extensions introduced in Goose v1.47.0.","The Goose extension is configured as:","However, `/authorize` is not the Keycloak authorization endpoint in this setup.","The actual Keycloak authorization endpoint is:"],"workaround":null,"fixed_in":null,"score":46.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"omacom/omarchy#14989","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":14989,"title":"[Shell] Qt 6.12 upgrade breaks Commons Color singleton (shadowed by QtQuick Color), causing black screen on menu and broken volume OSD","url":"https://github.com/omacom/omarchy/issues/14989","state":"closed","state_reason":"completed","labels":[],"created":"2026-10-11T07:37:35Z","updated":"2026-10-18T07:43:19Z","closed":"2026-10-18T07:43:19Z","reactions":4,"thumbs_up":4,"comments":6,"severity":"high","severity_reasons":["title:breakage"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"After upgrading to qt6-declarative 6.12.0-1 (and quickshell 0.3.2-1), opening the Omarchy menu (omarchy-menu toggle via SUPER + SPACE or SUPER + ESCAPE) causes a pitch-black screen with locked focus, and volume/brightness key adjustments…","versions":["6.12","6.12.0-1","0.3.2-1","4.0.4-1"],"error_snippets":["SUPER + ESCAPE","evaluates to","WlrKeyboardFocus.Exclusive","to crash with TypeErrors (","After upgrading to `qt6-declarative 6.12.0-1` (and `quickshell 0.3.2-1`), opening the Omarchy menu (`omarchy-menu toggle` via `SUPER + SPACE` or `SUPER + ESCAPE`) causes a pitch-black screen with locked focus, and volume"],"workaround":{"text":"Downgrade `qt6-` to `6.11.2` and `quickshell` to `0.3.1-1`.","source":"issue body","url":"https://github.com/omacom/omarchy/issues/14989"},"fixed_in":null,"score":46.1,"first_seen":"2026-10-19T18:38:33Z"},{"id":"github/copilot-cli#5085","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":5085,"title":"Rust rewrite regression on macOS: all native HTTPS requests fail with \"bad certificate format\" when com.apple.SecurityServer is not reachable (sandboxes)","url":"https://github.com/github/copilot-cli/issues/5085","state":"open","state_reason":null,"labels":["area:networking"],"created":"2026-10-08T15:15:15Z","updated":"2026-10-09T10:28:02Z","closed":null,"reactions":1,"thumbs_up":1,"comments":0,"severity":"high","severity_reasons":["title:breakage"],"topic":"self-host/headless","classes":["self-host/headless","install/update","always-on/daemon","auth/session","network"],"cosmetic_ui":false,"summary":"Starting with 1.0.93, Copilot CLI on macOS can't connect to its API when it runs inside a sandbox that denies the Mach service com.apple.SecurityServer. Sandboxes such as nono deny this service so the agent can't read keychain secrets…","versions":["1.0.93","1.0.92","1.0.94","1.0.95-2"],"error_snippets":["[ERROR] Error loading models: Error: error sending request for url (https://copilot-api.<host>/models): client error (Connect): bad certificate format","COPILOT_GITHUB_TOKEN=\"$(gh auth token)\" sandbox-exec -p \"$prof\" copilot --prefer-version 1.0.92 -p \"Reply with just: hi\" # works","COPILOT_GITHUB_TOKEN=\"$(gh auth token)\" sandbox-exec -p \"$prof\" copilot --prefer-version 1.0.93 -p \"Reply with just: hi\" # bad certificate format","sandbox-exec","(the legacy CSSM path) to evaluate certificates and returns"],"workaround":null,"fixed_in":null,"score":45.9,"first_seen":"2026-10-19T18:38:33Z"},{"id":"google-gemini/gemini-cli#29825","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":29825,"title":"MCP tool discovery ignores nextCursor and silently omits tools after the first page","url":"https://github.com/google-gemini/gemini-cli/issues/29825","state":"open","state_reason":null,"labels":["status/need-triage","area/agent"],"created":"2026-10-19T12:56:32Z","updated":"2026-10-20T09:35:58Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"medium","severity_reasons":["class:mcp/tools"],"topic":"mcp/tools","classes":["mcp/tools","self-host/headless","network"],"cosmetic_ui":false,"summary":"MCP tool discovery ignores nextCursor from a tools/list response. When a server advertises tools across multiple pages, Gemini CLI discovers only the first page, with no diagnostic that additional tools were omitted. On main at…","versions":["1.23.0","1.0.0"],"error_snippets":["enableEnvironmentVariableRedaction: true,","emitMcpDiagnostic: (...args) => diagnostics.push(args),","}, {}, { timeout: 1000 });","}, null, 2));","This leaves tools on later pages unavailable to the agent, even though the server is connected and has advertised that more results exist. The same discovery helper is used for initial discovery and tool-list refreshes. "],"workaround":null,"fixed_in":null,"score":45.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aaif-goose/goose#13003","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":13003,"title":"Image paths like `C:\\Users\\me\\shot.png` are no longer attached on Windows (regression since v1.39.0)","url":"https://github.com/aaif-goose/goose/issues/13003","state":"open","state_reason":null,"labels":[],"created":"2026-10-19T13:32:57Z","updated":"2026-10-19T14:57:08Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"high","severity_reasons":["title:breakage"],"topic":"install/update","classes":["install/update","self-host/headless","network"],"cosmetic_ui":false,"summary":"Describe the bug Since v1.39.0, goose no longer attaches an image when a message references it by a Windows path such as C:\\Users\\me\\Pictures\\shot.png (or C:/Users/me/Pictures/shot.png). The request sent to the provider only contains the…","versions":["1.39.0","1.38.0","1.55.0"],"error_snippets":[", so a drive-letter path is never considered. Before, each whitespace-separated word was checked with","2. Configure the LiteLLM provider with a vision-capable model (for example `gpt-6-luna`).","3. Run without extensions, so the model cannot open the file with a tool:","4. The image is not attached: the request sent to the provider only contains the text, so the model cannot see the image."],"workaround":null,"fixed_in":null,"score":45.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"google-gemini/gemini-cli#29555","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":29555,"title":"Unresponsive / Performance Issue","url":"https://github.com/google-gemini/gemini-cli/issues/29555","state":"open","state_reason":null,"labels":["area/core","status/bot-triaged","effort/small"],"created":"2026-09-29T15:15:48Z","updated":"2026-10-02T19:10:40Z","closed":null,"reactions":0,"thumbs_up":0,"comments":6,"severity":"medium","severity_reasons":["class:crash/hang"],"topic":"crash/hang","classes":["crash/hang"],"cosmetic_ui":false,"summary":"just looking up running extensions, this one is on top, taking quite a bit of loading time and... really not running at all anywhere i am currently accessing Extension version: 0.20.0 VS Code version: Code 1.139.1…","versions":["0.20.0","1.139.1","10.0.26200"],"error_snippets":["nes-ex-cf-euporie-03:31542868","just looking up running extensions, this one is on top, taking quite a bit of loading time and... really not running at all anywhere i am currently accessing","Extension version: 0.20.0","A/B Experiments"],"workaround":null,"fixed_in":null,"score":44.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"github/copilot-cli#5147","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":5147,"title":"/restart fails when the CLI was launched with --new-session-id","url":"https://github.com/github/copilot-cli/issues/5147","state":"open","state_reason":"reopened","labels":["area:sessions"],"created":"2026-10-15T01:21:02Z","updated":"2026-10-25T04:19:42Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"low","severity_reasons":["low engagement"],"topic":"mcp/tools","classes":["mcp/tools"],"cosmetic_ui":false,"summary":"Launching the CLI with --new-session-id and then running /restart fails every time: By the time this surfaces the foreground session has already been shut down, so the restart is not recoverable. copilot --new-session-id…","versions":["1.0.98"],"error_snippets":["error: the argument '--new-session-id <id>' cannot be used with '--session-id <id>'","--experimental","--no-experimental","Launching the CLI with `--new-session-id ` and then running `/restart` fails every time:","error: the argument '--new-session-id ' cannot be used with '--session-id '"],"workaround":null,"fixed_in":null,"score":44.5,"first_seen":"2026-10-25T09:11:37Z"},{"id":"aaif-goose/goose#12817","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":12817,"title":"Invalid cron updates remove the existing schedule timer, and an identical retry reports success","url":"https://github.com/aaif-goose/goose/issues/12817","state":"open","state_reason":null,"labels":[],"created":"2026-10-10T19:46:02Z","updated":"2026-10-15T18:46:22Z","closed":null,"reactions":0,"thumbs_up":0,"comments":2,"severity":"medium","severity_reasons":["class:always-on/daemon","class:install/update"],"topic":"always-on/daemon","classes":["always-on/daemon","install/update"],"cosmetic_ui":false,"summary":"Updating an existing schedule with an invalid cron expression disables its timer even though the update returns an error. The schedule remains in the in-memory job list, but its previous cron expression has already been replaced and its…","versions":[],"error_snippets":["with a valid recipe and cron expression","SchedulerError::CronParseError","This matters for unattended automations: a client can receive what looks like a normal validation error, leave the existing automation alone, and assume it will continue running at its previous cadence. Instead, that aut","The affected operation is `Scheduler::update_schedule` in the scheduler-enabled backend. The ACP handler for `_goose/unstable/schedules/update` delegates to this operation and maps a cron parsing error to an invalid-para","Use a scheduler-enabled backend and an existing, non-running schedule. The recipe contents and model configuration are not relevant to this update operation."],"workaround":null,"fixed_in":null,"score":44.4,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aaif-goose/goose#12970","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":12970,"title":"Support OAuth 2.0 / OIDC JWT authentication in LiteLLM provider","url":"https://github.com/aaif-goose/goose/issues/12970","state":"open","state_reason":null,"labels":[],"created":"2026-10-17T03:24:33Z","updated":"2026-10-17T03:59:59Z","closed":null,"reactions":1,"thumbs_up":1,"comments":0,"severity":"medium","severity_reasons":["class:auth/session"],"topic":"auth/session","classes":["auth/session"],"cosmetic_ui":false,"summary":"What problem would this solve? Enterprise deployments of LiteLLM proxy often enforce OAuth 2.0 / OIDC JWT authentication (enablejwtauth: true) backed by an identity provider (Keycloak, Okta, Entra ID / Azure AD, Auth0, Dex) rather than…","versions":["2.0"],"error_snippets":["permissions, refreshes expired tokens via"],"workaround":null,"fixed_in":null,"score":44.4,"first_seen":"2026-10-19T18:38:33Z"},{"id":"cline/cline#15084","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":15084,"title":"MCP servers never connect at startup - McpHub silently frozen (zero logs, no timeout, panel toggle dead)","url":"https://github.com/cline/cline/issues/15084","state":"open","state_reason":null,"labels":["VS Code"],"created":"2026-10-14T06:30:38Z","updated":"2026-10-14T06:50:46Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"medium","severity_reasons":["class:self-host/headless","class:crash/hang"],"topic":"self-host/headless","classes":["self-host/headless","crash/hang","mcp/tools","network","install/update"],"cosmetic_ui":false,"summary":"VSCode Extension 4.1.24 (also reproduced on 4.1.23 and 4.0.12) On this machine, Cline's MCP servers have never connected at startup — this is 100% reproducible, not intermittent. Every window reload ends the same way: Zero MCP-related…","versions":["4.1.24","4.1.23","4.0.12","4.1.22"],"error_snippets":["dist/extension.js","VSCode Extension","On this machine, Cline's MCP servers have **never** connected at startup — this is 100% reproducible, not intermittent. Every window reload ends the same way:","Code findings (shipped `dist/extension.js`):","1. McpHub constructor fires and forgets: `this.watchMcpSettingsFile(), this.initializeMcpServers()` — bare promises, no retry, no error surfacing. Both paths first `await getSettingsDirectoryPath()`; if that (or any late"],"workaround":null,"fixed_in":null,"score":44.3,"first_seen":"2026-10-19T18:38:33Z"},{"id":"kirodotdev/Kiro#11461","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":11461,"title":"WSL: device-code login prints the verification URL but never opens it, despite `wslview` being available","url":"https://github.com/kirodotdev/Kiro/issues/11461","state":"open","state_reason":null,"labels":["auth","theme:ssh-wsl","cli"],"created":"2026-09-18T05:36:43Z","updated":"2026-09-21T16:02:11Z","closed":null,"reactions":1,"thumbs_up":1,"comments":2,"severity":"medium","severity_reasons":["class:self-host/headless","class:auth/session"],"topic":"self-host/headless","classes":["self-host/headless","auth/session","install/update"],"cosmetic_ui":false,"summary":"Kiro CLI 2.22.0 WSL2 (Ubuntu 24.04), Windows 11 (build 26200) wslu/wslview installed; wslview opens the Windows default browser correctly Identity Center (org / --license pro) login When running kiro-cli login inside WSL, the CLI detects a…","versions":["2.22.0"],"error_snippets":["export BROWSER=wslview","When running `kiro-cli login` inside WSL, the CLI detects a \"remote\" environment and","Ctrl+click or copy/paste it every time — even though a working opener (`wslview`) is","`sensible-browser` first on `PATH` (each logs its args, then forwards to the real","\"tried to open and failed\" case; the open is simply never attempted on this path."],"workaround":null,"fixed_in":null,"score":44.1,"first_seen":"2026-10-19T18:38:33Z"},{"id":"omacom/omarchy#8628","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":8628,"title":"Lock screen: password field loses focus after display wake, recoverable only with a mouse","url":"https://github.com/omacom/omarchy/issues/8628","state":"closed","state_reason":"completed","labels":[],"created":"2026-08-27T16:52:03Z","updated":"2026-09-20T20:07:50Z","closed":"2026-09-20T20:07:50Z","reactions":5,"thumbs_up":5,"comments":5,"severity":"medium","severity_reasons":["class:always-on/daemon"],"topic":"always-on/daemon","classes":["always-on/daemon","crash/hang","mcp/tools","network"],"cosmetic_ui":false,"summary":"After the lock screen blanks the display, waking it does not restore focus to the password field. Keystrokes go nowhere. The only way to recover is to click with a mouse or tap the touchpad, which defeats keyboard-only operation. Omarchy…","versions":["4.0.1-1","0.56.2","0.3.1-1"],"error_snippets":["In `shell/plugins/lock/LockView.qml`, focus is asserted in exactly two places:","`root.wakeRequested()`, but an unfocused `TextInput` never receives key events,","shell loads it without QML errors and password PAM auth is unaffected.","An event-driven alternative would be to call `lockView.forcePasswordFocus()`"],"workaround":null,"fixed_in":null,"score":44.1,"first_seen":"2026-10-24T08:54:58Z"},{"id":"google-gemini/gemini-cli#29957","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":29957,"title":"GeminiCLI.com Feedback: olostep/olostep-mcp-server not indexed in the Extensions Gallery (gallery data stale since Oct 16)","url":"https://github.com/google-gemini/gemini-cli/issues/29957","state":"open","state_reason":null,"labels":["status/need-triage","area/extensions"],"created":"2026-10-26T09:34:13Z","updated":"2026-10-26T09:35:02Z","closed":null,"reactions":0,"thumbs_up":0,"comments":0,"severity":"medium","severity_reasons":["class:self-host/headless","class:mcp/tools"],"topic":"self-host/headless","classes":["self-host/headless","mcp/tools","install/update","network"],"cosmetic_ui":false,"summary":"https://geminicli.com/extensions/ Our extension https://github.com/olostep/olostep-mcp-server (olostep) is not listed in the Extensions Gallery, although it meets all documented requirements: Public repository gemini-cli-extension topic…","versions":["0.65.0"],"error_snippets":["gemini-cli-extension","gemini-extension.json","gemini extensions validate .","gemini extensions install https://github.com/olostep/olostep-mcp-server","https://geminicli.com/extensions/"],"workaround":null,"fixed_in":null,"score":44.0,"first_seen":"2026-10-26T10:12:02Z"},{"id":"github/copilot-cli#5256","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":5256,"title":"High CPU and repeated \"gh auth token\" failures when not signed in (worse without gh)","url":"https://github.com/github/copilot-cli/issues/5256","state":"open","state_reason":null,"labels":["triage"],"created":"2026-10-26T07:49:42Z","updated":"2026-10-26T08:16:04Z","closed":null,"reactions":0,"thumbs_up":0,"comments":0,"severity":"medium","severity_reasons":["class:crash/hang","class:auth/session"],"topic":"crash/hang","classes":["crash/hang","auth/session","install/update"],"cosmetic_ui":false,"summary":"When Copilot CLI runs without a usable GitHub credential, the background auth probe copilotruntime::auth::gh (i.e. gh auth token) is retried in a tight loop with no backoff. I hit this in a normal interactive session using a BYOK provider…","versions":["2.103.2"],"error_snippets":["[WARNING] [rust:copilot_runtime::auth::gh] Failed to run `gh auth token` {\"error\":\"program path has no file name\",\"program\":\"\"}","$env:COPILOT_HOME = \"C:\\temp\\cp-empty\" # empty dir => no credentials","2026-10-26T07:23:17.891Z [WARNING] [rust:copilot_runtime::auth::gh] Failed to run `gh auth token` {\"error\":\"program path has no file name\",\"program\":\"\"}","2026-10-26T07:23:17.896Z [WARNING] [rust:copilot_runtime::auth::gh] Failed to run `gh auth token` {\"error\":\"program path has no file name\",\"program\":\"\"}","2026-10-26T07:23:17.899Z [WARNING] [rust:copilot_runtime::auth::gh] Failed to run `gh auth token` {\"error\":\"program path has no file name\",\"program\":\"\"}"],"workaround":{"text":"Signing in (`/login`) or exporting `COPILOT_GITHUB_TOKEN` / a fine-grained PAT stops the loop; installing `gh` only lowers the severity.","source":"issue body","url":"https://github.com/github/copilot-cli/issues/5256"},"fixed_in":null,"score":44.0,"first_seen":"2026-10-26T10:12:02Z"},{"id":"continuedev/continue#13429","harness":"continue","harness_name":"Continue","repo":"continuedev/continue","number":13429,"title":"CLI 1.5.47: on a 401 the interactive CLI starts `npx -y mcp-remote` with no version pinned, in every OAuth profile we tested","url":"https://github.com/continuedev/continue/issues/13429","state":"open","state_reason":null,"labels":[],"created":"2026-10-26T07:03:04Z","updated":"2026-10-26T07:03:04Z","closed":null,"reactions":0,"thumbs_up":0,"comments":0,"severity":"medium","severity_reasons":["class:self-host/headless","class:auth/session"],"topic":"self-host/headless","classes":["self-host/headless","auth/session","mcp/tools","always-on/daemon","install/update"],"cosmetic_ui":false,"summary":"What happens @continuedev/cli 1.5.47 is the current release on npm. When isAuthError classifies an error as authentication, an interactive session falls back to npx -y mcp-remote in MCPService.ts. The predicate covers HTTP 401 and 405 and…","versions":["1.5.47","9.8","0.15.1","0.15.2"],"error_snippets":["classifies an error as authentication, an interactive session falls back to",", lines 441 to 474](https://github.com/continuedev/continue/blob/d3f60ba9dd3fb5bfd3c91d6fbb41ce1aa768db45/extensions/cli/src/services/MCPService.ts#L441-L474) and [","`@continuedev/cli` 1.5.47 is the current release on npm. When `isAuthError` classifies an error as authentication, an interactive session falls back to `npx -y mcp-remote ` in `MCPService.ts`. The predicate covers HTTP 4","The CLI's MCP transport carries only static headers and a Bearer `apiKey`. This bridge is therefore the CLI's only OAuth path. It came in with #8143. Source at the pinned commit: `MCPService.ts`, lines 441 to 474 and `mc","We ran CLI 1.5.47 on Windows against a local synthetic OAuth and MCP server, in four OAuth profiles. They were a public client without dynamic registration, a confidential client without it, a Client ID Metadata Document"],"workaround":null,"fixed_in":null,"score":44.0,"first_seen":"2026-10-26T10:12:02Z"},{"id":"openclaw/openclaw#167345","harness":"openclaw","harness_name":"OpenClaw","repo":"openclaw/openclaw","number":167345,"title":"Let a turn continue after a mid-stream `WebSocket closed 1000` on the ChatGPT Responses transport","url":"https://github.com/openclaw/openclaw/issues/167345","state":"closed","state_reason":"completed","labels":["no-stale","P1","clawsweeper:fix-shape-clear","clawsweeper:queueable-fix","clawsweeper:source-repro","impact:auth-provider","issue-rating: 🦞 diamond lobster"],"created":"2026-10-08T17:05:35Z","updated":"2026-10-12T22:36:56Z","closed":"2026-10-12T22:36:56Z","reactions":3,"thumbs_up":2,"comments":3,"severity":"high","severity_reasons":["label:p1"],"topic":"auth/session","classes":["auth/session","network"],"cosmetic_ui":false,"summary":"On 2026.9.8, a ChatGPT Responses WebSocket that closes with code 1000 after dispatch ends the run with \"Agent run failed\". Neither the same-model retry nor the settled-batch continuation from #130721 runs, and the session idles until the…","versions":["2026.9.8"],"error_snippets":["[openai-transport] ChatGPT Responses stream terminated provider=openai api=openai-chatgpt-responses model=gpt-6.1-sol transport=auto elapsedMs=23841 stopReason=error failureKind=transport","[agent/embedded] embedded run agent end: … isError=true model=gpt-6.1-sol provider=openai error=⚠️ Agent run failed (model: openai/gpt-6.1-sol). rawError=WebSocket closed 1000","[agent/embedded] incomplete turn detected: … stopReason=error hasLastAssistant=yes hasCurrentAttemptAssistant=yes … replaySafe=no …","TERMINAL_ASSISTANT_ERROR_CODES","isRetryableAssistantError"],"workaround":null,"fixed_in":null,"score":43.9,"first_seen":"2026-10-26T10:12:02Z"},{"id":"cline/cline#15270","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":15270,"title":"Billing: accidental annual upgrade — $77.22 credit balance never refunded, support unresponsive since Aug 26 (5 emails)","url":"https://github.com/cline/cline/issues/15270","state":"closed","state_reason":"completed","labels":[],"created":"2026-10-22T12:14:18Z","updated":"2026-10-22T14:06:18Z","closed":"2026-10-22T14:05:58Z","reactions":0,"thumbs_up":0,"comments":3,"severity":"medium","severity_reasons":["class:install/update","class:crash/hang"],"topic":"install/update","classes":["install/update","crash/hang","auth/session"],"cosmetic_ui":false,"summary":"Cloud Platform N/A - ClinePass web billing issue I am using a beta version of Cline Cloud Platform — ClinePass web billing (app.cline.bot) I am reporting a ClinePass billing issue and requesting a refund of a USD 77.22 credit balance that…","versions":["77.22","5.30","77.23","2.38"],"error_snippets":["Timeline (all invoices under account email already provided to support by email):","Support contact record — I emailed support@cline.bot five times (Aug 26, Aug 26, Aug 27, Aug 31, and Oct 22, 2026) with my receipt and invoice details. I have never received any reply, not even an acknowledgement.","This is the same flow reported in #14390, #15260 and #13983: an instant annual charge, the difference parked as account credit instead of a card refund, and no response from support. I note that one reporter in #14390 di","3. Please refund the USD 77.22 credit balance to my original payment method (Mastercard ending in 8054). I no longer have an active subscription, so there is nothing this credit can be applied to","My account email, receipts and invoice details have already been sent privately to support@cline.bot five times; I can provide them again through a private channel."],"workaround":null,"fixed_in":null,"score":43.9,"first_seen":"2026-10-26T10:12:02Z"},{"id":"github/copilot-cli#4971","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":4971,"title":"Every hour I get Authorization error. Your credentials may be expired or invalid","url":"https://github.com/github/copilot-cli/issues/4971","state":"open","state_reason":null,"labels":["area:authentication"],"created":"2026-09-26T00:52:02Z","updated":"2026-10-04T07:04:17Z","closed":null,"reactions":2,"thumbs_up":2,"comments":3,"severity":"critical","severity_reasons":["title:data-loss/security"],"topic":"auth/session","classes":["auth/session"],"cosmetic_ui":false,"summary":"Every hour or so I'm getting Authorization error. Your credentials may be expired or invalid. (Request ID: ) Run /login to sign in again. Running /login does finish successfully but does not remedy the issue. Nor does mcp reload. GitHub…","versions":["1.0.88"],"error_snippets":["Authorization error. Your credentials may be expired or invalid. (Request ID: <AN ID>) Run /login to sign in again.","Every hour or so I'm getting `Authorization error. Your credentials may be expired or invalid. (Request ID: ) Run /login to sign in again.`","Just use the CLI for long enough. One of the messages will randomly trigger this."],"workaround":null,"fixed_in":null,"score":43.7,"first_seen":"2026-10-19T18:38:33Z"},{"id":"anomalyco/opencode#49033","harness":"opencode","harness_name":"OpenCode","repo":"anomalyco/opencode","number":49033,"title":"Models get stuck on \"Thinking\" after a few hours","url":"https://github.com/anomalyco/opencode/issues/49033","state":"open","state_reason":null,"labels":[],"created":"2026-09-14T19:20:37Z","updated":"2026-09-14T19:56:00Z","closed":null,"reactions":2,"thumbs_up":2,"comments":3,"severity":"medium","severity_reasons":["class:crash/hang"],"topic":"crash/hang","classes":["crash/hang"],"cosmetic_ui":false,"summary":"After using OpenCode for a few hours, any model I try gets stuck indefinitely on \"Thinking\". There is no error message and the request never finishes. Switching to another model does not help. I have not reached any usage or rate limits…","versions":[],"error_snippets":["There is no error message and the request never finishes. Switching to another model does not help.","Expected behavior","The model should respond normally or show an error."],"workaround":null,"fixed_in":null,"score":43.7,"first_seen":"2026-10-19T18:38:33Z"},{"id":"github/copilot-cli#5008","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":5008,"title":"Startup error \"Failed to read model provider attribution: Error: Not authenticated\" in 1.0.89","url":"https://github.com/github/copilot-cli/issues/5008","state":"closed","state_reason":"completed","labels":["area:authentication","area:models"],"created":"2026-09-30T03:11:02Z","updated":"2026-10-04T11:41:10Z","closed":"2026-10-04T11:41:10Z","reactions":5,"thumbs_up":5,"comments":7,"severity":"medium","severity_reasons":["class:auth/session"],"topic":"auth/session","classes":["auth/session"],"cosmetic_ui":false,"summary":"Since 1.0.89, every new interactive session shows this error twice right after startup: Failed to read model provider attribution: Error: Not authenticated Sign-in completes about 3 seconds later and chat/model requests work normally. It…","versions":["1.0.89","1.0.85-1.0.88"],"error_snippets":["Since 1.0.89, every new interactive session shows this error twice right after startup:","Failed to read model provider attribution: Error: Not authenticated","Sign-in completes about 3 seconds later and chat/model requests work normally. It looks like a startup race: the new provider-attribution code (`model-picker` and `model-provider-labels`) calls `model.list()` before auth","1. Authenticate via the `COPILOT_GITHUB_TOKEN` environment variable (OAuth token; no accounts stored via `/login`).","3. The error appears twice in the timeline immediately after startup."],"workaround":null,"fixed_in":null,"score":43.3,"first_seen":"2026-10-19T18:38:33Z"},{"id":"cline/cline#14700","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":14700,"title":"Cline Remote: helper daemon leak causes permanent connection failure + unhandled OOM crash","url":"https://github.com/cline/cline/issues/14700","state":"open","state_reason":null,"labels":[],"created":"2026-09-30T09:55:32Z","updated":"2026-09-30T16:18:34Z","closed":null,"reactions":0,"thumbs_up":0,"comments":2,"severity":"high","severity_reasons":["title:breakage"],"topic":"always-on/daemon","classes":["always-on/daemon","self-host/headless","crash/hang","install/update","network"],"cosmetic_ui":false,"summary":"0.0.39 (previously 0.0.32 → 0.0.37) I am using a beta version of Cline Connecting to a remote environment fails with \"Unable to connect to SSH host\". The remote helper daemon (--cline-hub-daemon) is leaked and never reaped after failed…","versions":["0.0.39","0.0.32","0.0.37"],"error_snippets":["Unable to connect to SSH host","Failed to run remote command /home/<user>/.cline/remote/cline-remote-helper-linux-arm64-2d97a33d88ce0590 (exit 1):","{\"time\":\"2026-09-30T09:14:35.711Z\",\"platform_version\":\"0.0.39\", \"command\":\"connect_remote_environment\",\"severity\":\"error\",\"handled\":true, \"error_message\":\"Failed to run remote command .../cline-remote-helper-linux-arm64-","PID 4172614 PPID 1 etimes 156345 RSS 435MB ...-904dded7ce0844fe --cline-hub-daemon","PID 2583176 PPID 1 etimes 3595 RSS 395MB ...-904dded7ce0844fe --cline-hub-daemon"],"workaround":null,"fixed_in":null,"score":43.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aaif-goose/goose#12266","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":12266,"title":"ACP: MCP extension startup failures never reach the ACP client — session/new succeeds, no session/update, warning only in goose's log file","url":"https://github.com/aaif-goose/goose/issues/12266","state":"open","state_reason":null,"labels":[],"created":"2026-09-20T23:30:20Z","updated":"2026-10-24T04:45:23Z","closed":null,"reactions":0,"thumbs_up":0,"comments":2,"severity":"medium","severity_reasons":["class:install/update","class:mcp/tools"],"topic":"install/update","classes":["install/update","mcp/tools","self-host/headless","network"],"cosmetic_ui":false,"summary":"When an MCP server passed in ACP session/new fails to start, the ACP client is never told. The session/new response is a normal success, no session/update is sent, and nothing appears on stdout or stderr. The host has no way to learn that…","versions":["1.51.0","0.17.0","2.0"],"error_snippets":["WARN goose::agents::agent | Failed to load extension broken-mcp: failed to initialize MCP client:","Send message error Transport [rmcp::transport::worker::WorkerTransport<…streamable…>] … ConnectionRefused","threading.Thread(target=lambda: [errs.append(l.rstrip()) for l in p.stderr], daemon=True).start()","print(\"error in session/new?\", \"error\" in r)","print(\"stderr lines:\", len(errs))"],"workaround":null,"fixed_in":null,"score":42.9,"first_seen":"2026-10-24T08:54:58Z"},{"id":"omacom/omarchy#11458","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":11458,"title":"Hard freeze shortly after resuming from s2idle on Intel Ice Lake (i915) with kernel 7.2.3-arch1-3 (ASUS ZenBook UX325JA)","url":"https://github.com/omacom/omarchy/issues/11458","state":"open","state_reason":null,"labels":["bug","compatibility"],"created":"2026-09-12T12:48:46Z","updated":"2026-10-14T00:43:56Z","closed":null,"reactions":1,"thumbs_up":1,"comments":3,"severity":"high","severity_reasons":["title:breakage"],"topic":"crash/hang","classes":["crash/hang","always-on/daemon","install/update","mcp/tools","network"],"cosmetic_ui":false,"summary":"On Omarchy 4.0.3-1 with kernel 7.2.3-arch1-3, this laptop intermittently hard-freezes ~20-50 s after a successful s2idle resume. Suspend and resume both complete and log normally (PM: suspend entry and PM: suspend exit); the machine then…","versions":["4.0.3-1","7.2.3-arch1","7.2.3","7.1.9-arch1"],"error_snippets":["Sep 12 07:30:43 omarchy kernel: PM: suspend entry (s2idle)","Sep 12 07:30:48 omarchy kernel: PM: suspend exit","Sep 10 07:06:19 omarchy kernel: PM: suspend entry (s2idle)","Sep 10 07:06:23 omarchy kernel: PM: suspend exit","PM: suspend entry"],"workaround":{"text":"Pin/downgrade to kernel 7.1.9-arch1-2 (or use `linux-lts`) until a fixed kernel is available. - Try `i915.enable_psr=0` / `i915.enable_dc=0`, since the failure is in the post-resume display re-initialisation path. - Try `mem_sleep_default=deep` to move off s2idle.","source":"issue body","url":"https://github.com/omacom/omarchy/issues/11458"},"fixed_in":null,"score":42.4,"first_seen":"2026-10-25T09:11:37Z"},{"id":"omacom/omarchy#13394","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":13394,"title":"Lenovo ThinkPad does not shut down properly and slow startup","url":"https://github.com/omacom/omarchy/issues/13394","state":"open","state_reason":null,"labels":["bug","compatibility"],"created":"2026-09-27T10:07:20Z","updated":"2026-10-13T16:35:35Z","closed":null,"reactions":1,"thumbs_up":0,"comments":10,"severity":"low","severity_reasons":["low engagement"],"topic":"other","classes":[],"cosmetic_ui":false,"summary":"Laptop: Lenovo ThinkPad T460, CPU: Intel(R) Core(TM) i7-6600U (4)z, Graphics: Intel HD Graphics 520 / Skylake-U GT2, GPU driver: i915, Omarchy: 4.0.4-1 When I shut down the laptop: The screen turns off. The power button light turns off…","versions":["4.0.4-1"],"error_snippets":["Laptop: Lenovo ThinkPad T460, CPU: Intel(R) Core(TM) i7-6600U (4)z, Graphics: Intel HD Graphics 520 / Skylake-U GT2, GPU driver: i915, Omarchy: 4.0.4-1"],"workaround":null,"fixed_in":null,"score":42.2,"first_seen":"2026-10-19T18:38:33Z"},{"id":"kirodotdev/Kiro#5457","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":5457,"title":"Context usage always shows 0% and crashes with \"unexpected error\" when context is full","url":"https://github.com/kirodotdev/Kiro/issues/5457","state":"closed","state_reason":"not_planned","labels":["theme:context-limit-issue"],"created":"2026-02-06T03:22:36Z","updated":"2026-09-24T13:25:47Z","closed":"2026-09-24T13:25:38Z","reactions":6,"thumbs_up":6,"comments":6,"severity":"high","severity_reasons":["title:breakage"],"topic":"crash/hang","classes":["crash/hang"],"cosmetic_ui":false,"summary":"The context usage indicator in the chat panel always displays 0%, regardless of how many files have been read or how long the conversation is. When the context window is actually full (after reading many files or having a long…","versions":["0.9.2"],"error_snippets":["When the context window is actually full (after reading many files or having a long conversation), Kiro throws a generic error:","\"An unexpected error occurred while processing your input.\"","5、Eventually the agent responds with \"An unexpected error occurred while processing your input\"","6、Impact: Users cannot predict when to start a new session, leading to lost work and frustration.","When the limit is hit, show a clear error message like \"Context window is full. Please start a new chat session.\" instead of a generic unexpected error"],"workaround":null,"fixed_in":null,"score":42.1,"first_seen":"2026-10-19T18:38:33Z"},{"id":"kirodotdev/Kiro#9495","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":9495,"title":"Kiro – Previous session chat history lost/inaccessible after 1.0.0 update","url":"https://github.com/kirodotdev/Kiro/issues/9495","state":"closed","state_reason":"completed","labels":[],"created":"2026-06-17T18:13:45Z","updated":"2026-10-16T16:03:10Z","closed":"2026-07-01T22:12:23Z","reactions":5,"thumbs_up":5,"comments":8,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"Product: Kiro IDE Version: 1.0.0 (updated morning of June 17, 2026) OS: Windows Trigger: Occurred immediately after updating Kiro to version 1.0.0 and enabling the Agent Focus experimental feature. Issue: Previous session chat history is…","versions":["1.0.0"],"error_snippets":["Trigger: Occurred immediately after updating Kiro to version 1.0.0 and enabling the Agent Focus experimental feature.","Issue: Previous session chat history is no longer accessible. The session directory (C:\\Users\\pcuser\\.kiro\\sessions\\67dc5747e37490bc\\) still exists on disk with 38+ sub-session folders, but the conversations cannot be lo","Expected behavior: Previous session history remains accessible after update. User can resume or reference prior conversations.","Actual behavior: All prior chat history appears lost from the UI. Session data exists on disk but cannot be loaded. New sessions start with no context from previous work.","User must manually rebuild context each session by re-reading files"],"workaround":null,"fixed_in":null,"score":42.1,"first_seen":"2026-10-25T09:11:37Z"},{"id":"aaif-goose/goose#13122","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":13122,"title":"Add The Agent Forge to the extensions directory (remote MCP, no auth)","url":"https://github.com/aaif-goose/goose/issues/13122","state":"closed","state_reason":"completed","labels":[],"created":"2026-10-24T11:15:06Z","updated":"2026-10-25T23:59:07Z","closed":"2026-10-25T23:59:07Z","reactions":0,"thumbs_up":0,"comments":1,"severity":"medium","severity_reasons":["class:self-host/headless","class:auth/session"],"topic":"self-host/headless","classes":["self-host/headless","auth/session","mcp/tools","install/update"],"cosmetic_ui":false,"summary":"What problem would this solve? Goose users can't easily see how AI agents behave on everyday business jobs with planted traps, test their own agent against those traps, or check a skill, MCP server or package before trusting it. The Agent…","versions":[],"error_snippets":["Goose users can't easily see how AI agents behave on everyday business jobs with planted traps, test their own agent against those traps, or check a skill, MCP server or package before trusting it. The Agent Forge is a f","The Agent Forge appears in the extensions directory, so a Goose user can add it in one step.","Verification: the JSON stays valid, the docs site builds, and the extension connects in Goose Desktop and lists its tools. Docs: https://theagentforge.io/connect"],"workaround":null,"fixed_in":null,"score":41.6,"first_seen":"2026-10-25T09:11:37Z"},{"id":"google-gemini/gemini-cli#29591","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":29591,"title":"⚠  High memory usage detected: 7.28 GB. If you experience a crash, please file a bug report by running /bug","url":"https://github.com/google-gemini/gemini-cli/issues/29591","state":"open","state_reason":null,"labels":["area/core","status/bot-triaged","effort/large"],"created":"2026-10-01T14:51:34Z","updated":"2026-10-12T10:50:24Z","closed":null,"reactions":0,"thumbs_up":0,"comments":4,"severity":"high","severity_reasons":["title:breakage"],"topic":"crash/hang","classes":["crash/hang","auth/session"],"cosmetic_ui":false,"summary":"Crash happened using model gemini-3-flash-preview trying to do the ingest phase of a LLM wiki. I uploaded 2 80-slides pdfs and one short md and asked LLM for ingest. The LLM thought for 18-min reaching the end of the free-tokens , switched…","versions":["7.28","7912.6","7921.2","7830.7"],"error_snippets":["[19779:0x99d3000] 1709109 ms: Mark-Compact 7912.6 (7921.2) -> 7830.7 (7842.8) MB, pooled: 0.0 MB, 25.29 / 0.00 ms (average mu = 0.991, current mu = 0.989) allocation failure; scavenge might not succeedemini-3-flash-previ","FATAL ERROR: Reached heap limit Allocation failed - JavaScript heap out of memory","1: 0x8f364b node::OOMErrorHandler(char const*, v8::OOMDetails const&) [/home/alessia/.nvm/versions/node/v25.6.1/bin/node]","⚠ High memory usage detected: 7.28 GB. If you experience a crash, please file a bug report by running /bug","`⚠ High memory usage detected: 7.28 GB. If you experience a crash, please file a bug report by running /bug`"],"workaround":null,"fixed_in":null,"score":41.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aaif-goose/goose#12881","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":12881,"title":"Expose goose context token so user can select it manually if auto set tokens size limit is invalid","url":"https://github.com/aaif-goose/goose/issues/12881","state":"open","state_reason":null,"labels":[],"created":"2026-10-13T21:02:24Z","updated":"2026-10-18T16:58:18Z","closed":null,"reactions":0,"thumbs_up":0,"comments":2,"severity":"medium","severity_reasons":["class:auth/session"],"topic":"auth/session","classes":["auth/session"],"cosmetic_ui":false,"summary":"What problem would this solve? This will prevent context of the model from being set incorrectly and allow user to correct the model context size. What would a good outcome look like? when clicking on the model context token, user can…","versions":[],"error_snippets":[],"workaround":null,"fixed_in":null,"score":41.4,"first_seen":"2026-10-19T18:38:33Z"},{"id":"anomalyco/opencode#51928","harness":"opencode","harness_name":"OpenCode","repo":"anomalyco/opencode","number":51928,"title":"providers: GitHub Copilot model sync never runs after login, no models listed","url":"https://github.com/anomalyco/opencode/issues/51928","state":"closed","state_reason":"duplicate","labels":[],"created":"2026-09-28T18:57:16Z","updated":"2026-10-17T20:11:12Z","closed":"2026-10-17T20:11:12Z","reactions":4,"thumbs_up":4,"comments":4,"severity":"medium","severity_reasons":["class:auth/session"],"topic":"auth/session","classes":["auth/session","always-on/daemon","self-host/headless","install/update","mcp/tools","network"],"cosmetic_ui":false,"summary":"GitHub Copilot OAuth login succeeds and the account is stored, but no Copilot models ever appear in /models or opencode models. The model-list sync is never attempted after login: the server log contains no sync success or failure for the…","versions":["2.0.18"],"error_snippets":["2026-09-25 level=WARN message=\"failed to sync GitHub Copilot models\" cause=\"TimeoutError: The operation timed out.\"","2026-09-26 level=WARN message=\"failed to sync GitHub Copilot models\" cause=\"TimeoutError: The operation timed out.\"","TimeoutError","Provider not found","GitHub Copilot OAuth login succeeds and the account is stored, but no Copilot models ever appear in `/models` or `opencode models`. The model-list sync is never attempted after login: the server log contains no sync succ"],"workaround":null,"fixed_in":null,"score":41.3,"first_seen":"2026-10-19T18:38:33Z"},{"id":"github/copilot-cli#4977","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":4977,"title":"Bundled ripgrep aborts with jemalloc 'Unsupported system page size' on 16KB-page ARM64 kernels (Asahi Linux)","url":"https://github.com/github/copilot-cli/issues/4977","state":"open","state_reason":null,"labels":["area:platform-linux","area:installation","area:tools"],"created":"2026-09-27T15:22:33Z","updated":"2026-10-25T19:33:05Z","closed":null,"reactions":0,"thumbs_up":0,"comments":2,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update","crash/hang"],"cosmetic_ui":false,"summary":"The ripgrep binary bundled with the Copilot CLI ( /ripgrep/bin/linux-arm64/rg) statically links jemalloc built assuming a 4KB page size. On a kernel configured with a 16KB page size — the default on Asahi Linux (Linux-on-Apple-Silicon) —…","versions":["1.0.88","1.0.83","1.0.87"],"error_snippets":["$ echo hello | ~/.cache/copilot/pkg/linux-arm64/1.0.88/ripgrep/bin/linux-arm64/rg hello","memory allocation of 160 bytes failed","Aborted (exit code 134)",".../ripgrep/bin/linux-arm64/rg: ELF 64-bit LSB executable, ARM aarch64, version 1 (SYSV), statically linked, stripped","at each cached version's"],"workaround":{"text":"Replacing the bundled `rg` at each cached version's `ripgrep/bin/linux-arm64/rg` with the system's dynamically-linked `rg` resolves the crash locally, but is undone on every CLI update since it re-fetches the broken bundle.","source":"issue body","url":"https://github.com/github/copilot-cli/issues/4977"},"fixed_in":null,"score":41.1,"first_seen":"2026-10-26T10:12:02Z"},{"id":"anomalyco/opencode#51434","harness":"opencode","harness_name":"OpenCode","repo":"anomalyco/opencode","number":51434,"title":"opencode-go/deepseek-v4.1-flash randomly rejects requests with 400/403","url":"https://github.com/anomalyco/opencode/issues/51434","state":"closed","state_reason":"not_planned","labels":[],"created":"2026-09-26T02:32:13Z","updated":"2026-10-11T19:38:26Z","closed":"2026-10-11T19:38:26Z","reactions":4,"thumbs_up":4,"comments":5,"severity":"medium","severity_reasons":["class:auth/session"],"topic":"auth/session","classes":["auth/session","always-on/daemon","mcp/tools"],"cosmetic_ui":false,"summary":"I'm hitting random failures with deepseek-v4.1-flash on the opencode-go provider. The session just stops with: Bad Request: {\"model\":\"deepseek-v4.1-flash\"} Sometimes I also see it come back as a 403 with error code: 1010 when I replay the…","versions":["1.18.32"],"error_snippets":["error code: 1010","I'm hitting random failures with deepseek-v4.1-flash on the opencode-go provider. The session just stops with:","Sometimes I also see it come back as a 403 with `error code: 1010` when I replay the same request by hand.","I pulled the message history for the failed turn straight out of `~/.local/share/opencode/opencode.db` and sent the exact same messages array to `https://opencode.ai/zen/go/v1/chat/completions` with a valid key and sessi","4. Open `~/.local/share/opencode/opencode.db` (sqlite), pull the messages/parts for the failed assistant message, and rebuild the same `messages[]` array."],"workaround":null,"fixed_in":null,"score":41.1,"first_seen":"2026-10-19T18:38:33Z"},{"id":"omacom/omarchy#11744","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":11744,"title":"Idle wake is dead after a lock, and a clamshell-disabled internal panel survives resume unreconciled","url":"https://github.com/omacom/omarchy/issues/11744","state":"open","state_reason":null,"labels":["bug"],"created":"2026-09-13T22:32:05Z","updated":"2026-10-11T23:57:43Z","closed":null,"reactions":1,"thumbs_up":0,"comments":4,"severity":"medium","severity_reasons":["class:always-on/daemon"],"topic":"always-on/daemon","classes":["always-on/daemon","mcp/tools"],"cosmetic_ui":false,"summary":"Updated after the discussion below. The blank-to-link-drop mechanism v1b3m describes is confirmed on this machine too and belongs with #9078 and #10691, so I have split it out of this report. What remains here are the two findings that the…","versions":[],"error_snippets":["19:23:50 eDP-1 disabled=true (clamshell: lid closed + external monitor active)","19:24:17 eDP-1 re-enabled (watcher poll, 3 s later)","hl.dsp.dpms({ action = \"enable\" })","amdgpu ... DMCUB error","`lockSystem` in `plugins/services/idle/Service.qml` clears `idledThisCycle` at line 77 immediately before triggering the lock at line 80. `handleActiveSignal` then returns early on `!root.idledThisCycle` at line 159. So "],"workaround":null,"fixed_in":null,"score":41.0,"first_seen":"2026-10-26T10:12:02Z"},{"id":"anomalyco/opencode#49992","harness":"opencode","harness_name":"OpenCode","repo":"anomalyco/opencode","number":49992,"title":"VSCode plugin broken","url":"https://github.com/anomalyco/opencode/issues/49992","state":"closed","state_reason":"completed","labels":[],"created":"2026-09-19T15:03:49Z","updated":"2026-10-12T18:58:39Z","closed":"2026-10-12T18:58:39Z","reactions":6,"thumbs_up":6,"comments":5,"severity":"high","severity_reasons":["title:breakage"],"topic":"mcp/tools","classes":["mcp/tools"],"cosmetic_ui":false,"summary":"On start it use opencode --port 16955 Unrecognized flag: --port in command opencode but correct syntax is opencode serve --port 16955 opencode v2.0.10 Install opencode & vscode plugin Use button with black square on white \"Open opencode in…","versions":["2.0.10"],"error_snippets":["3. Opened terminal with command and error"],"workaround":null,"fixed_in":null,"score":40.9,"first_seen":"2026-10-19T18:38:33Z"},{"id":"omacom/omarchy#7596","harness":"omarchy","harness_name":"Omarchy","repo":"omacom/omarchy","number":7596,"title":"Quickshell unable to restart after latest update on Edge branch","url":"https://github.com/omacom/omarchy/issues/7596","state":"closed","state_reason":"completed","labels":["bug"],"created":"2026-08-20T09:26:35Z","updated":"2026-09-14T17:34:21Z","closed":"2026-08-21T11:19:13Z","reactions":5,"thumbs_up":5,"comments":7,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"i5-10400F, RTX 2080 Ti, Omarchy 4.0.0.r1787.g9455496-1 quickshell: symbol lookup error: quickshell: undefined symbol: ZN23QUntypedPropertyBindingC1EP23QPropertyBindingPrivate, version Qt6PRIVATEAPI 1) Tried reboot 2) Re-tried Omarchy…","versions":[],"error_snippets":["quickshell: symbol lookup error: quickshell: undefined symbol: _ZN23QUntypedPropertyBindingC1EP23QPropertyBindingPrivate, version Qt_6_PRIVATE_API"],"workaround":null,"fixed_in":null,"score":40.9,"first_seen":"2026-10-26T10:12:02Z"},{"id":"cline/cline#14836","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":14836,"title":"Model can't see MCP tools added either manually or via marketplace","url":"https://github.com/cline/cline/issues/14836","state":"open","state_reason":null,"labels":[],"created":"2026-10-05T10:48:35Z","updated":"2026-10-05T19:11:56Z","closed":null,"reactions":0,"thumbs_up":0,"comments":3,"severity":"medium","severity_reasons":["class:mcp/tools"],"topic":"mcp/tools","classes":["mcp/tools","install/update"],"cosmetic_ui":false,"summary":"I am using a beta version of Cline I'm having problems getting agents to see installed MCPs in Cline Desktop on Windows. Whether they are manually installed or installed via the marketplace, when I ask the model to list the tools available…","versions":["0.0.43"],"error_snippets":[],"workaround":null,"fixed_in":null,"score":40.8,"first_seen":"2026-10-19T18:38:33Z"},{"id":"cline/cline#14701","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":14701,"title":"Desktop 0.0.38: bundled aarch64 remote helper fails --remote-hub-ensure with Bun SyntaxError: Invalid character: '\\0' (previous build's helper works on the same host)","url":"https://github.com/cline/cline/issues/14701","state":"open","state_reason":null,"labels":[],"created":"2026-09-30T12:37:01Z","updated":"2026-10-08T10:25:32Z","closed":null,"reactions":0,"thumbs_up":0,"comments":4,"severity":"medium","severity_reasons":["class:self-host/headless"],"topic":"self-host/headless","classes":["self-host/headless","always-on/daemon"],"cosmetic_ui":false,"summary":"Cline Desktop 0.0.38 (core 0.0.88) cannot connect to a Linux aarch64 SSH remote. The bundled cline-remote-helper-aarch64-unknown-linux-gnu (sha256 93444d160eed1dc9…) exits 1 on --remote-hub-ensure with a Bun SyntaxError: Invalid character…","versions":["0.0.38","0.0.88"],"error_snippets":["$ $H --remote-hub-ensure --cwd /home/ubuntu --discovery-path /home/ubuntu/.cline/data/remote/probe.json </dev/null","2788 | `),process.exitCode=1});","SyntaxError: Invalid character: '\\0'","at emit (node:events:100:22)","--remote-hub-ensure --cwd /home/ubuntu --discovery-path /home/ubuntu/.cline/data/remote/probe-old.json </dev/null"],"workaround":{"text":"`mkdir -p ~/.cline/remote-helpers ; scp :~/.cline/remote/cline-remote-helper-linux-arm64-8b51183e615e7442 \\`","source":"issue body","url":"https://github.com/cline/cline/issues/14701"},"fixed_in":null,"score":40.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aaif-goose/goose#12016","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":12016,"title":"MCP OAuth re-authorizes on every new session after transient refresh failures or refresh rotation races","url":"https://github.com/aaif-goose/goose/issues/12016","state":"open","state_reason":null,"labels":[],"created":"2026-09-11T16:46:57Z","updated":"2026-09-24T15:37:52Z","closed":null,"reactions":1,"thumbs_up":1,"comments":1,"severity":"medium","severity_reasons":["class:auth/session","class:mcp/tools"],"topic":"auth/session","classes":["auth/session","mcp/tools","self-host/headless","network"],"cosmetic_ui":false,"summary":"Describe the bug MCP servers that authenticate via OAuth demand a fresh interactive browser authorization on (nearly) every new goose session, even though valid credentials were previously stored. Once it starts happening, the stored…","versions":[],"error_snippets":["TokenRefreshFailed","per working directory, plus the CLI). With authorization servers that rotate refresh tokens on every use (e.g. Atlassian/Trello, Composio), the process that loses the refresh race receives","token_response: null","[OAuth:...] Token refresh failed ... clearing stored credentials and falling back to browser auth","AuthError::TokenRefreshRejected"],"workaround":null,"fixed_in":null,"score":40.3,"first_seen":"2026-10-19T18:38:33Z"},{"id":"google-gemini/gemini-cli#29793","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":29793,"title":"GeminiCLI.com Feedback: ANTEVO-CH/antevo-mcp missing from the Extensions Gallery (topic + root manifest since Sept)","url":"https://github.com/google-gemini/gemini-cli/issues/29793","state":"open","state_reason":null,"labels":["area/extensions","status/bot-triaged","effort/large"],"created":"2026-10-16T23:16:43Z","updated":"2026-10-17T01:27:02Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"medium","severity_reasons":["class:mcp/tools"],"topic":"mcp/tools","classes":["mcp/tools","self-host/headless","install/update"],"cosmetic_ui":false,"summary":"https://geminicli.com/extensions/ The extension in https://github.com/ANTEVO-CH/antevo-mcp (manifest name antevo) is missing from the Extensions Gallery. It is not in https://geminicli.com/extensions.json (last-modified 16 Oct 2026 02:36…","versions":["0.64.0"],"error_snippets":[") is missing from the Extensions Gallery. It is not in","gemini-cli-extension","search/repositories?q=topic:gemini-cli-extension+org:ANTEVO-CH","gemini-extension.json","npx @google/gemini-cli@0.64.0 extensions validate ."],"workaround":null,"fixed_in":null,"score":40.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"github/copilot-cli#5165","harness":"copilot-cli","harness_name":"GitHub Copilot CLI","repo":"github/copilot-cli","number":5165,"title":"1.0.98: Copilot Enterprise account routed to api.business.githubcopilot.com — 401 on model catalog, 'Authentication failed' despite valid token","url":"https://github.com/github/copilot-cli/issues/5165","state":"open","state_reason":null,"labels":["area:authentication","area:enterprise"],"created":"2026-10-16T10:22:29Z","updated":"2026-10-17T05:23:27Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"medium","severity_reasons":["class:auth/session"],"topic":"auth/session","classes":["auth/session","self-host/headless","install/update","mcp/tools"],"cosmetic_ui":false,"summary":"Copilot CLI 1.0.98 (installed via official gh.io/copilot-install script, Linux x64, headless — no system keychain, token stored in plaintext ~/.copilot/config.json) Auth via copilot login --device-code on github.com (not a GHEC…","versions":["1.0.98"],"error_snippets":["Error: Authentication failed (Request ID: ...)","Your GitHub token may be invalid, expired, or lacking the required permissions.","[ERROR] worker quit with fatal: Transport channel closed, when Client(OAuthChallenge { status_code: 401 ... })","[WARNING] MCP OAuth authorization failed; reporting the server as needing auth {\"server\":\"github-mcp-server\",\"error\":\"Browser-based OAuth required for https://api.business.githubcopilot.com/mcp/readonly\"}","[ERROR] Error loading models: Error: 401 \"Unauthorized\""],"workaround":null,"fixed_in":null,"score":40.0,"first_seen":"2026-10-23T09:23:24Z"},{"id":"google-gemini/gemini-cli#29693","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":29693,"title":"Make the 60 s API headers timeout configurable (long-thinking requests are dropped and replayed)","url":"https://github.com/google-gemini/gemini-cli/issues/29693","state":"open","state_reason":null,"labels":["area/core","status/bot-triaged","effort/small"],"created":"2026-10-09T09:55:09Z","updated":"2026-10-09T13:25:28Z","closed":null,"reactions":0,"thumbs_up":0,"comments":2,"severity":"medium","severity_reasons":["class:network"],"topic":"network","classes":["network","self-host/headless","install/update","crash/hang","auth/session"],"cosmetic_ui":false,"summary":"A setting (or environment variable) for the API request headers timeout, honoured with every auth type, including Vertex AI. For example model.requestTimeoutMs in settings.json, or GEMINIAPIHEADERSTIMEOUTMS. Since #26191 the global undici…","versions":["0.62.0"],"error_snippets":["model.requestTimeoutMs","GEMINI_API_HEADERS_TIMEOUT_MS","headersTimeout","UND_ERR_HEADERS_TIMEOUT","updateGlobalFetchTimeouts"],"workaround":{"text":"Our workaround is to preload a module (`NODE_OPTIONS=--import`) that re-creates the global `Agent` with a longer `headersTimeout` when the CLI installs it.","source":"issue body","url":"https://github.com/google-gemini/gemini-cli/issues/29693"},"fixed_in":null,"score":39.8,"first_seen":"2026-10-19T18:38:33Z"},{"id":"cline/cline#15107","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":15107,"title":"Cline CLI auto-exits ~5 seconds after startup","url":"https://github.com/cline/cline/issues/15107","state":"open","state_reason":null,"labels":["CLI"],"created":"2026-10-15T03:15:14Z","updated":"2026-10-15T03:31:26Z","closed":null,"reactions":0,"thumbs_up":0,"comments":2,"severity":"low","severity_reasons":["low engagement"],"topic":"always-on/daemon","classes":["always-on/daemon","install/update","network"],"cosmetic_ui":false,"summary":"I am using a beta version of Cline Cline CLI starts up normally — the ASCII logo, \"What can I do for you?\", the input box, model line (MiMo-V2.6-Flash (free) (medium)), git status (ai agent (master) | 51 files +1375) and \"Auto-approve all…","versions":["3.0.70"],"error_snippets":["Cline CLI starts up normally — the ASCII logo, \"What can I do for you?\", the input box, model line (MiMo-V2.6-Flash (free) (medium)), git status (ai agent (master) | 51 files +1375) and \"Auto-approve all enabled (Shift+T","The screenshot shows the CLI in Windows Terminal on Windows 11, fully rendered and idle, ~5 seconds before it exits.","4. The process exits on its own"],"workaround":null,"fixed_in":null,"score":39.8,"first_seen":"2026-10-19T18:38:33Z"},{"id":"kirodotdev/Kiro#11369","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":11369,"title":"kiro-cli v3: hook_update reports a failed hook command as completed (root cause)","url":"https://github.com/kirodotdev/Kiro/issues/11369","state":"open","state_reason":null,"labels":["hooks","pending-maintainer-response","cli"],"created":"2026-09-13T19:19:52Z","updated":"2026-09-17T21:40:14Z","closed":null,"reactions":1,"thumbs_up":1,"comments":3,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update","self-host/headless"],"cosmetic_ui":false,"summary":"Debian 13 (trixie), x8664. The mechanism is OS-independent: the status is a string literal in the agent runtime's own JavaScript, written before the command result is looked at. kiro-cli 2.21.4, @kiro/agent 0.63.3, engine v3 (KAS). Not a…","versions":["2.21.4","0.63.3"],"error_snippets":["let d=await this.deps.commandAction.execute(u,c,{cwd:r.cwd,signal:r.signal,timeoutSeconds:u.timeoutSeconds}),p=Hmr(r.trigger,d.exitCode),","function Mk(e,t){t.name&&e.emit({type:\"AgentExecutionAction\",executionId:e.executionId,actionId:Sja(),actionState:\"Success\",actionType:\"ContextualHookInvoked\",input:{name:t.name,hookId:t.id,hookActionType:Lmr(t.actionKin","\"action\": {\"type\": \"command\", \"command\": \"echo PROBE_FAIL_STDERR 1>&2; exit 3\"}},","0.63.3, engine v3 (KAS). Not a v2 report: the Rust in-process agent loop never loads","hooks: {enabled: true, v2: true}"],"workaround":null,"fixed_in":null,"score":39.7,"first_seen":"2026-10-21T09:24:21Z"},{"id":"cline/cline#14435","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":14435,"title":"Cline Desktop Beta 0.0.34 stopped working after update","url":"https://github.com/cline/cline/issues/14435","state":"open","state_reason":null,"labels":["beta","Windows","Desktop App"],"created":"2026-09-23T04:22:53Z","updated":"2026-09-28T20:53:30Z","closed":null,"reactions":1,"thumbs_up":1,"comments":3,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update","crash/hang"],"cosmetic_ui":false,"summary":"I am using a beta version of Cline After updating to version 0.0.34 Beta, the application stopped working properly. Before the update everything was fine, but now the software does not start or respond as expected. I did not change…","versions":["0.0.34"],"error_snippets":["After updating to version 0.0.34 Beta, the application stopped working properly. Before the update everything was fine, but now the software does not start or respond as expected. I did not change anything in my system, ","Error message: \"Desktop backend transport did not become ready within 90000ms (Tauri invoke failed for get_desktop_backend_endpoint: desktop backend endpoint not ready)\"","The app fails to respond or throws an unexpected error","App is stuck on loading screen, cannot access About menu. Error displayed on the interface: \"Desktop backend transport did not become ready within 90000ms (Tauri invoke failed for get_desktop_backend_endpoint: desktop ba"],"workaround":null,"fixed_in":null,"score":39.7,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aws/amazon-q-developer-cli#2768","harness":"amazon-q-cli","harness_name":"Amazon Q Developer CLI","repo":"aws/amazon-q-developer-cli","number":2768,"title":"Amazon Q CLI not loading MCP servers despite correct configuration","url":"https://github.com/aws/amazon-q-developer-cli/issues/2768","state":"closed","state_reason":"not_planned","labels":[],"created":"2025-09-03T18:10:19Z","updated":"2026-09-29T16:46:46Z","closed":"2026-09-29T16:46:45Z","reactions":4,"thumbs_up":4,"comments":0,"severity":"medium","severity_reasons":["class:self-host/headless","class:mcp/tools"],"topic":"self-host/headless","classes":["self-host/headless","mcp/tools","install/update","auth/session","network"],"cosmetic_ui":false,"summary":"Can't use MCP anymore. Tool Permission executebash trust read-only commands fsread trusted fswrite not trusted introspect trusted reportissue trusted useaws trust read-only commands Servers still loading version = \"1.15.0\" hash =…","versions":["1.15.0","16.00","5.9","24.7.0"],"error_snippets":["Can't use MCP anymore.","[environment]","✔ Everything looks good!","• ~/.aws/amazonq/mcp-servers.json (563 bytes)","• ~/.aws/amazonq/mcpAdmin/mcp-state.json → {\"enabled\":true}"],"workaround":null,"fixed_in":null,"score":39.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"cline/cline#15180","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":15180,"title":"cli has progressively longer \"hangs\"","url":"https://github.com/cline/cline/issues/15180","state":"open","state_reason":null,"labels":["CLI"],"created":"2026-10-18T21:26:02Z","updated":"2026-10-18T21:26:11Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"high","severity_reasons":["title:breakage"],"topic":"crash/hang","classes":["crash/hang"],"cosmetic_ui":false,"summary":"I am using a beta version of Cline I'm using a large-context thinking LLM, which outputs a lot of code-blocks and patches. When starting a new session, cline is snappy / responsive. But as the context / history gets longer, the TUI begins…","versions":["3.0.73"],"error_snippets":["This resolves after a certain delay – which increases with bigger history: compaction does not help, exiting cline and restarting into the existing session does not help. When it has resolved, new content from the LLM is","It does not happen on every output, but seems to be related to Markdown / Syntax highlighting, as you can sometimes uncoloured output (of things that were coloured before) and notice how it becomes coloured only after a "],"workaround":null,"fixed_in":null,"score":39.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"cline/cline#14570","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":14570,"title":"CLI hangs indefinitely in non-interactive (-c) mode on Linux ARM64","url":"https://github.com/cline/cline/issues/14570","state":"open","state_reason":null,"labels":["CLI"],"created":"2026-09-27T02:39:21Z","updated":"2026-10-12T01:19:22Z","closed":null,"reactions":0,"thumbs_up":0,"comments":4,"severity":"high","severity_reasons":["title:breakage"],"topic":"crash/hang","classes":["crash/hang","self-host/headless","install/update","auth/session","network"],"cosmetic_ui":false,"summary":"I am using a beta version of Cline The Cline CLI (v3.0.65, also reproduced on 3.0.38) hangs indefinitely at startup when run in non-interactive/headless mode with the -c flag. The interactive TUI mode (cline without flags) works fine on…","versions":["3.0.65","3.0.38"],"error_snippets":["4. Observe: the process hangs indefinitely. The log file (`~/.cline/data/logs/cline.log`) shows only the initial \"CLI run started\" entry and nothing further, even after 10+ minutes."],"workaround":null,"fixed_in":null,"score":39.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"google-gemini/gemini-cli#29592","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":29592,"title":"macOS PTY master leak persists in 0.62.0 after completed shell commands (token-free repro)","url":"https://github.com/google-gemini/gemini-cli/issues/29592","state":"open","state_reason":null,"labels":["area/core","status/bot-triaged","effort/large"],"created":"2026-10-01T15:14:48Z","updated":"2026-10-08T13:36:27Z","closed":null,"reactions":0,"thumbs_up":0,"comments":2,"severity":"medium","severity_reasons":["class:crash/hang","class:auth/session"],"topic":"crash/hang","classes":["crash/hang","auth/session","always-on/daemon","install/update","network"],"cosmetic_ui":false,"summary":"Gemini CLI 0.62.0 on macOS retains one distinct /dev/ptmx master per completed ordinary shell command in a minimal token-free reproduction. This looks like a recurrence of #15945 after #25079. It exhausts the host PTY limit and prevents…","versions":["0.62.0"],"error_snippets":["// Update this filename from the CLI entry-point imports for a different build.","const entry = 'chunk-MLY4WQFO.js';","const {ShellExecutionService} = await import(pathToFileURL(path.join(bundle, entry)));","const count = () => spawnSync('lsof', ['-nP', '-a', '-p', String(process.pid), '/dev/ptmx'], {encoding:'utf8'}).stdout.split('\\n').filter(x => x.includes('/dev/ptmx')).length;","const handle = await ShellExecutionService.execute("],"workaround":null,"fixed_in":null,"score":38.9,"first_seen":"2026-10-19T18:38:33Z"},{"id":"cline/cline#15077","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":15077,"title":"[Bug]: OpenCode provider strips provider prefix (causing fallback to server default) & hangs on external workspace permissions","url":"https://github.com/cline/cline/issues/15077","state":"open","state_reason":null,"labels":["VS Code"],"created":"2026-10-14T05:17:59Z","updated":"2026-10-14T05:18:08Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"high","severity_reasons":["title:breakage"],"topic":"self-host/headless","classes":["self-host/headless","crash/hang"],"cosmetic_ui":false,"summary":"VSCode Extension I am using a beta version of Cline When configuring the OpenCode provider in Cline Desktop and selecting any model from the dropdown (such as LongCat 2.5 Preview Free): Incorrect Model Execution: Even though…","versions":["0.0.46","2.5","2.5-preview","2.6-flash"],"error_snippets":[", or edits freeze indefinitely showing","VSCode Extension","1. **Incorrect Model Execution**: Even though `longcat-2.5-preview-free` is selected in Cline, the assistant replies using the server's default model (`mimo-v2.6-flash-free`). The selected model ID is silently omitted fr","2. **Infinite Hang on File Tools**: When working on any workspace located outside the user's home directory (e.g. `C:\\Users\\Public\\...`), tool calls like `read`, `glob`, or edits freeze indefinitely showing `⚪ Reading fi"],"workaround":null,"fixed_in":null,"score":38.8,"first_seen":"2026-10-19T18:38:33Z"},{"id":"anomalyco/opencode#52548","harness":"opencode","harness_name":"OpenCode","repo":"anomalyco/opencode","number":52548,"title":"`opencode auth login` crashes on Linux without xdg-open","url":"https://github.com/anomalyco/opencode/issues/52548","state":"closed","state_reason":"completed","labels":[],"created":"2026-10-01T18:01:09Z","updated":"2026-10-16T13:42:52Z","closed":"2026-10-16T13:42:52Z","reactions":2,"thumbs_up":2,"comments":3,"severity":"high","severity_reasons":["title:breakage"],"topic":"crash/hang","classes":["crash/hang","auth/session","self-host/headless"],"cosmetic_ui":false,"summary":"On a Linux box without xdg-utils (SSH, containers), opencode auth login prints the authorization URL and then exits with Error: Executable not found in $PATH: \"xdg-open\" and a stack trace, so the login can't be finished. openUrl in…","versions":["2.0.21"],"error_snippets":["● Enter code: XXXX-XXXX","Error: Executable not found in $PATH: \"xdg-open\"","Effect.promise(...).pipe(Effect.ignore)","Effect.promise","Effect.ignore"],"workaround":null,"fixed_in":null,"score":38.6,"first_seen":"2026-10-19T18:38:33Z"},{"id":"cline/cline#14622","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":14622,"title":"CLI hub mode: MCP server processes are never stopped after /clear or quitting; they accumulate until the hub daemon exits","url":"https://github.com/cline/cline/issues/14622","state":"open","state_reason":null,"labels":["CLI","beta"],"created":"2026-09-28T21:07:13Z","updated":"2026-09-30T16:25:09Z","closed":null,"reactions":0,"thumbs_up":0,"comments":2,"severity":"medium","severity_reasons":["class:always-on/daemon","class:self-host/headless"],"topic":"always-on/daemon","classes":["always-on/daemon","self-host/headless","mcp/tools","network"],"cosmetic_ui":false,"summary":"I am using a beta version of Cline In the CLI's hub mode, a session's stdio MCP server processes are never stopped when you leave that session, whether by /clear or by quitting the TUI. Each new session starts its own set of MCP servers…","versions":["3.0.65"],"error_snippets":["or by quitting the TUI. Each new session starts its own set of MCP servers, and the old ones stay alive under the","@cline/core/dist/hub/daemon/entry.js","), a failed session, a finished **non-interactive** run (","In the CLI's hub mode, a session's stdio MCP server processes are never stopped when you leave that session, whether by `/clear` or by quitting the TUI. Each new session starts its own set of MCP servers, and the old one","4. `pgrep -af ` shows one server process per session ever created since the hub started, all still running under the `cline --cline-hub-daemon` process."],"workaround":null,"fixed_in":null,"score":38.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"cline/cline#14606","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":14606,"title":"Orphaned hub daemon is reused after sidecar restart, failing every prompt with \"Hub connection closed (code=1006)\"","url":"https://github.com/cline/cline/issues/14606","state":"open","state_reason":null,"labels":[],"created":"2026-09-28T11:19:44Z","updated":"2026-10-04T13:05:06Z","closed":null,"reactions":0,"thumbs_up":0,"comments":4,"severity":"medium","severity_reasons":["class:always-on/daemon"],"topic":"always-on/daemon","classes":["always-on/daemon","network"],"cosmetic_ui":false,"summary":"I am using a beta version of Cline Leaving the Cline Desktop App running overnight causes the WebSocket connection between the sidecar and the local hub daemon to drop immediately on the next launch. Every prompt then fails with a hard…","versions":["0.0.37","24.6","43.779","43.781"],"error_snippets":["The run failed: Hub connection closed (code=1006, reason=Connection ended)","10:13:44.041 ERROR Hub connection closed (code=1006) <- dies ~0.3s later","10:13:46.877 ERROR Hub connection closed (code=1006)","10:13:48.546 client.register","10:13:49.074 ERROR Hub connection closed (code=1006)"],"workaround":null,"fixed_in":null,"score":38.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aaif-goose/goose#12084","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":12084,"title":"Make extension state and MCP server lifecycle consistent","url":"https://github.com/aaif-goose/goose/issues/12084","state":"open","state_reason":null,"labels":[],"created":"2026-09-15T18:15:24Z","updated":"2026-09-24T05:49:54Z","closed":null,"reactions":0,"thumbs_up":0,"comments":3,"severity":"medium","severity_reasons":["class:self-host/headless","class:mcp/tools"],"topic":"self-host/headless","classes":["self-host/headless","mcp/tools","crash/hang"],"cosmetic_ui":false,"summary":"What problem would this solve? The backend currently treats the extensions selected for a session and the MCP clients running for them as one mutable ExtensionManager. The same component resolves configuration and secrets, starts…","versions":[],"error_snippets":["ExtensionManager","The backend currently treats the extensions selected for a session and the MCP clients running for them as one mutable `ExtensionManager`. The same component resolves configuration and secrets, starts transports, owns li","Extension behavior is consistent across sessions, ACP connections, agent paths, and code mode. A change is visible everywhere the same session is used, session execution context is preserved, capability discovery does no","The backend is easier to evolve without exposing connection and lifecycle details to its callers."],"workaround":null,"fixed_in":null,"score":38.2,"first_seen":"2026-10-25T09:11:37Z"},{"id":"cline/cline#14094","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":14094,"title":"VS Code server extension slows down then crashes","url":"https://github.com/cline/cline/issues/14094","state":"open","state_reason":null,"labels":["VS Code"],"created":"2026-09-13T06:37:16Z","updated":"2026-09-14T11:58:09Z","closed":null,"reactions":0,"thumbs_up":0,"comments":3,"severity":"high","severity_reasons":["title:breakage"],"topic":"self-host/headless","classes":["self-host/headless","crash/hang","auth/session"],"cosmetic_ui":false,"summary":"VSCode Extension I am using a beta version of Cline Happens while displaying the stream of thinking tokens. Display slows down, then stops (but I know the model is actually producing tokens cause it is internally hosted), then bottom…","versions":["4.1.17"],"error_snippets":["VSCode Extension","Happens while displaying the stream of thinking tokens. Display slows down, then stops (but I know the model is actually producing tokens cause it is internally hosted), then bottom status bar of VS Code server (in the b"],"workaround":null,"fixed_in":null,"score":38.2,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aws/amazon-q-developer-cli#3932","harness":"amazon-q-cli","harness_name":"Amazon Q Developer CLI","repo":"aws/amazon-q-developer-cli","number":3932,"title":"Kiro CLI 2.33.1 on macOS bundles Node.js 22.22.2. Node.js 22.23.4 includes security fixes released October 21, 2026. Please update the bundled runtime.","url":"https://github.com/aws/amazon-q-developer-cli/issues/3932","state":"open","state_reason":null,"labels":[],"created":"2026-10-23T11:43:46Z","updated":"2026-10-23T11:43:46Z","closed":null,"reactions":0,"thumbs_up":0,"comments":0,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"I have run q doctor in the affected terminal session I have run q restart and replicated the issue again macOS 26.7.1 (25G313) Kiro points to the updated Version of note Kiro points to the unsafe Version of note","versions":["2.33.1","36.00","5.9","3.13"],"error_snippets":["[environment]"],"workaround":null,"fixed_in":null,"score":38.0,"first_seen":"2026-10-24T08:54:58Z"},{"id":"cline/cline#15091","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":15091,"title":"Desktop 0.0.45: text-only session with ~23 MB accumulated tool results makes EVERY send fail — \"Hub connection closed (code=1006)\" / hub-daemon.log: \"Received too big message\"","url":"https://github.com/cline/cline/issues/15091","state":"open","state_reason":null,"labels":[],"created":"2026-10-14T12:10:20Z","updated":"2026-10-14T12:10:25Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"medium","severity_reasons":["class:always-on/daemon"],"topic":"always-on/daemon","classes":["always-on/daemon","mcp/tools","network"],"cosmetic_ui":false,"summary":"0.0.45 (extensionversion 0.0.45, clinetype: desktop, platformversion 0.0.45) A long-running agent session (hundreds of tool calls; tool results carry multi-KB JSON including full Python source; no images at all) accumulated a 23.1 MB…","versions":["0.0.45","23.1","0.0.37","0.0.26"],"error_snippets":["The run failed: Hub connection closed (code=1006, reason=Connection ended)","The session is permanently unusable — every subsequent message fails the same way. A fresh session works fine.","hub-daemon.log contains **10 occurrences** of \"Received too big message\" since 2026-10-10, each correlating with a 22–23 MB session file.","{\"time\":\"2026-10-14T06:12:50.013Z\",\"event\":\"sdk.error\",\"command\":\"chat_session_command\",","\"operation\":\"command.execute\",\"severity\":\"error\",\"handled\":true,"],"workaround":null,"fixed_in":null,"score":38.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"cline/cline#14733","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":14733,"title":"github mcp not works","url":"https://github.com/cline/cline/issues/14733","state":"open","state_reason":null,"labels":[],"created":"2026-10-01T08:43:59Z","updated":"2026-10-01T10:18:32Z","closed":null,"reactions":0,"thumbs_up":0,"comments":3,"severity":"medium","severity_reasons":["class:mcp/tools"],"topic":"mcp/tools","classes":["mcp/tools","install/update","auth/session"],"cosmetic_ui":false,"summary":"I am using a beta version of Cline i'm install github mcp via marketplace and it cant authorize on github why it not able to authorize in github like cloudflare-api mcp ? install github mcp via cline marketplace try to use github mcp via…","versions":["0.0.40"],"error_snippets":["3. 403 error in prompt response"],"workaround":null,"fixed_in":null,"score":37.9,"first_seen":"2026-10-19T18:38:33Z"},{"id":"cline/cline#15032","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":15032,"title":"Desktop app quits mid-session with no crash report; orphaned hub daemon persists","url":"https://github.com/cline/cline/issues/15032","state":"open","state_reason":null,"labels":[],"created":"2026-10-12T17:11:14Z","updated":"2026-10-12T17:11:18Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"high","severity_reasons":["title:breakage"],"topic":"always-on/daemon","classes":["always-on/daemon","crash/hang"],"cosmetic_ui":false,"summary":"I am using a beta version of Cline macOS (Cline Desktop app, not the VS Code extension) Cline version: Mac: 16 GB RAM, up 5 days, load average ~4.5 MCP servers enabled: 5 (aiemaily, narrateai, zapier, browseros, buildin) The app closes…","versions":["0.0.45","4.5","5.3"],"error_snippets":["log show --predicate 'eventMessage CONTAINS \"jetsam\"'","The app closes without warning, both mid-task and shortly after a task completes. The window disappears and the app leaves the dock. No error is shown to the user.","2. Not memory pressure. `memory_pressure -Q` reports 33% free (~5.3 GB). `log show --predicate 'eventMessage CONTAINS \"jetsam\"'` shows only idle-exit kills of system daemons — no memory-driven kills. `memorystatus_availa","3. Not automatic termination. `defaults read /Applications/Cline.app/Contents/Info.plist NSSupportsAutomaticTermination` returns \"does not exist\" — the app does not opt into macOS automatic termination, so the OS should ","6. Orphaned hub daemon recurs. After a clean quit + pkill of all sidecar processes, a new `code-sidecar --cline-hub-daemon --cwd / --host 127.0.0.1 --port 25463 --pathname /hub` reappears with parent PID 1 (launchd) and "],"workaround":{"text":"`pkill -f \"cline-hub-daemon\"; sleep 1; open -a Cline` before each session.","source":"issue body","url":"https://github.com/cline/cline/issues/15032"},"fixed_in":null,"score":37.8,"first_seen":"2026-10-19T18:38:33Z"},{"id":"anomalyco/opencode#49139","harness":"opencode","harness_name":"OpenCode","repo":"anomalyco/opencode","number":49139,"title":"[Bug] Mistral provider (zai-glm-5-2) fails with AI_TypeValidationError on streamed tool calls (@ai-sdk/mistral dependency update needed)","url":"https://github.com/anomalyco/opencode/issues/49139","state":"closed","state_reason":"not_planned","labels":[],"created":"2026-09-15T07:00:58Z","updated":"2026-10-11T19:14:35Z","closed":"2026-10-11T19:14:35Z","reactions":3,"thumbs_up":3,"comments":3,"severity":"medium","severity_reasons":["class:install/update","class:mcp/tools"],"topic":"install/update","classes":["install/update","mcp/tools"],"cosmetic_ui":false,"summary":"OpenCode Version: 1.18.29 Model / Provider: Mistral (mistral/zai-glm-5-2) When using Mistral models (specifically zai-glm-5-2) with tool calling enabled, streaming responses crash with an AITypeValidationError. The error trace indicates…","versions":["1.18.29"],"error_snippets":["AI_TypeValidationError: Type validation failed:","Value: {\"id\":\"34ab1513a87f422fbc31733708620b96\",\"object\":\"chat.completion.chunk\",\"created\":1789112573,\"model\":\"zai-glm-5-2\",\"choices\":[{\"index\":0,\"delta\":{\"tool_calls\":[{\"type\":\"function\",\"function\":{\"name\":\"\",\"arguments","Error message: [","\"expected\": \"string\",","\"message\": \"Invalid input: expected string, received undefined\""],"workaround":null,"fixed_in":null,"score":37.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aaif-goose/goose#12337","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":12337,"title":"Define origin isolation for MCP App guest content","url":"https://github.com/aaif-goose/goose/issues/12337","state":"open","state_reason":null,"labels":[],"created":"2026-09-21T14:14:25Z","updated":"2026-10-15T16:26:58Z","closed":null,"reactions":0,"thumbs_up":0,"comments":5,"severity":"medium","severity_reasons":["class:mcp/tools"],"topic":"mcp/tools","classes":["mcp/tools"],"cosmetic_ui":false,"summary":"What problem would this solve? MCP App guest content is served from a stable loopback web origin. Without host-enforced origin isolation, separately supplied apps can share browser origin state and may interact with host cookies or storage…","versions":[],"error_snippets":["MCP App guest content is served from a stable loopback web origin. Without host-enforced origin isolation, separately supplied apps can share browser origin state and may interact with host cookies or storage in ways use","Each app's guest content is isolated from local host applications and from other apps while the supported MCP Apps feature set continues to work.","Enforce an opaque-origin sandbox policy without same-origin privileges; serve apps from distinct, unpredictable origins; use separate browser storage partitions or sessions per app; or explicitly define required support "],"workaround":null,"fixed_in":null,"score":37.0,"first_seen":"2026-10-25T09:11:37Z"},{"id":"anomalyco/opencode#49777","harness":"opencode","harness_name":"OpenCode","repo":"anomalyco/opencode","number":49777,"title":"tui: /btw crashes when rendering answer dialog","url":"https://github.com/anomalyco/opencode/issues/49777","state":"closed","state_reason":"completed","labels":[],"created":"2026-09-18T13:38:03Z","updated":"2026-09-27T14:14:11Z","closed":"2026-09-19T14:09:58Z","reactions":3,"thumbs_up":3,"comments":6,"severity":"high","severity_reasons":["title:breakage"],"topic":"crash/hang","classes":["crash/hang","mcp/tools"],"cosmetic_ui":false,"summary":"Stop letting @thdxr push :) Running the built-in /btw slash command crashes the TUI after the generated answer is returned. The answer dialog attempts to access PluginProvider outside its context boundary. opencode version: 2.0.8 Shell…","versions":["2.0.8","1.3.1"],"error_snippets":["Error: PluginProvider is missing","1. Open an existing session in the TUI.","2. Enter `/btw `."],"workaround":null,"fixed_in":null,"score":36.9,"first_seen":"2026-10-19T18:38:33Z"},{"id":"kirodotdev/Kiro#11801","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":11801,"title":"## CRASH #10737 — 2026-10-03","url":"https://github.com/kirodotdev/Kiro/issues/11801","state":"open","state_reason":null,"labels":["os: windows","extensions","pending-maintainer-response","theme:unexpected-error"],"created":"2026-10-06T04:41:28Z","updated":"2026-10-13T15:18:15Z","closed":null,"reactions":0,"thumbs_up":0,"comments":2,"severity":"high","severity_reasons":["title:breakage"],"topic":"crash/hang","classes":["crash/hang","always-on/daemon","mcp/tools","network"],"cosmetic_ui":false,"summary":"STATUS: FURIOUS. ABSOLUTELY LIVID. Kiro decided to shit the bed mid-session Extension host processes spawning like rabbits (PIDs 45736, 46296) Grammar scope overwrites flying everywhere (SQL, Java, YAML all fighting) Deprecation warnings…","versions":["1.2.4"],"error_snippets":["workbench.desktop.main.js:442 ERR [Extension Host] (node:45736) [DEP0169] DeprecationWarning: url.parse()","workbench.desktop.main.js:442 ERR [Extension Host] (node:46296) [DEP0169] DeprecationWarning: url.parse()","mermaid-GHXKKRXX-BLVsJ6Xc.js:522 [tiptap warn]: Duplicate extension names found: ['slashCommand']","WARN [redhat.vscode-xml]: Cannot register 'redhat.telemetry.enabled'. Already registered.","WARN [redhat.vscode-yaml]: Cannot register 'redhat.telemetry.enabled'. Already registered."],"workaround":null,"fixed_in":null,"score":36.8,"first_seen":"2026-10-19T18:38:33Z"},{"id":"Aider-AI/aider#5908","harness":"aider","harness_name":"Aider","repo":"Aider-AI/aider","number":5908,"title":"Uncaught FileNotFoundError in <frozen posixpath> line 412","url":"https://github.com/Aider-AI/aider/issues/5908","state":"closed","state_reason":"completed","labels":[],"created":"2026-10-25T16:03:51Z","updated":"2026-10-26T00:08:47Z","closed":"2026-10-26T00:08:47Z","reactions":0,"thumbs_up":0,"comments":1,"severity":"medium","severity_reasons":["class:crash/hang"],"topic":"crash/hang","classes":["crash/hang"],"cosmetic_ui":false,"summary":"Aider version: 0.86.2 Python version: 3.12.15 Python implementation: CPython Virtual environment: Yes Git version: git version 2.50.1 (Apple Git-155) An uncaught exception occurred:","versions":["0.86.2","3.12.15","2.50.1"],"error_snippets":["Traceback (most recent call last):","sys.exit(main())","File \"io.py\", line 543, in get_input","File \"<frozen posixpath>\", line 412, in abspath","FileNotFoundError: [Errno 2] No such file or directory"],"workaround":null,"fixed_in":null,"score":36.3,"first_seen":"2026-10-26T10:12:02Z"},{"id":"kirodotdev/Kiro#11645","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":11645,"title":"CLI 2.24.1: KIRO_API_KEY works with agent-engine v1/v2 but V3/KAS returns 403 invalid bearer token","url":"https://github.com/kirodotdev/Kiro/issues/11645","state":"open","state_reason":null,"labels":["os: linux","auth","pending-maintainer-response","cli"],"created":"2026-09-28T10:20:41Z","updated":"2026-09-28T17:32:29Z","closed":null,"reactions":0,"thumbs_up":0,"comments":4,"severity":"medium","severity_reasons":["class:auth/session"],"topic":"auth/session","classes":["auth/session"],"cosmetic_ui":false,"summary":"CLI 2.24.1: KIROAPIKEY works with agent-engine v1/v2 but V3/KAS returns 403 invalid bearer token export KIROAPIKEY='ksk...' kiro-cli chat --no-interactive --agent-engine v1 \"Reply with OK\" kiro-cli chat --no-interactive --agent-engine v2…","versions":["2.24.1"],"error_snippets":["CLI 2.24.1: KIRO_API_KEY works with agent-engine v1/v2 but V3/KAS returns 403 invalid bearer token","export KIRO_API_KEY='ksk_...'","kiro-cli chat --no-interactive --agent-engine v1 \"Reply with OK\"","kiro-cli chat --no-interactive --agent-engine v2 \"Reply with OK\"","kiro-cli chat --no-interactive --agent-engine v3 \"Reply with OK\""],"workaround":{"text":"export KIRO_API_KEY='ksk_...'","source":"issue body","url":"https://github.com/kirodotdev/Kiro/issues/11645"},"fixed_in":null,"score":36.1,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aaif-goose/goose#11158","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":11158,"title":"[Bug / OAuth]: Goose Desktop v1.45.0: MCP Server that use query parameters fails at connection time in with rmcp \"unexpected server response\" error","url":"https://github.com/aaif-goose/goose/issues/11158","state":"open","state_reason":null,"labels":[],"created":"2026-08-12T03:48:42Z","updated":"2026-10-15T16:26:30Z","closed":null,"reactions":0,"thumbs_up":0,"comments":2,"severity":"medium","severity_reasons":["class:self-host/headless","class:auth/session"],"topic":"self-host/headless","classes":["self-host/headless","auth/session","mcp/tools"],"cosmetic_ui":false,"summary":"Describe the bug Cannot connect to the Datadog MCP server via Oauth Expected behavior When connecting to the Datadog MCP server ( https://mcp.datadoghq.com/api/unstable/mcp-server/mcp?toolsets=core with a custom query string parameter to…","versions":["1.45.0","1.44.0"],"error_snippets":["failed to initialize MCP client: Send message error Transport [rmcp::transport::worker::WorkerTransport<rmcp::transport::streamable_http_client::StreamableHttpClientWorker<reqwest::async_impl::client::Client>>] error: un","Cannot connect to the Datadog MCP server via Oauth","Error message when clicking copy","failed to initialize MCP client: Send message error Transport [rmcp::transport::worker::WorkerTransport >] error: unexpected server response: HTTP 401 Unauthorized: {\"errors\":[\"Unauthorized\"]}, when send initialize reque"],"workaround":{"text":"Downgrading to `1.44.0` restores connectivity. https://github.com/aaif-goose/goose/releases#release-v1.44.0","source":"issue body","url":"https://github.com/aaif-goose/goose/issues/11158"},"fixed_in":null,"score":35.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"cline/cline#14751","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":14751,"title":"Cline Desktop: [Linux]: Configured section will not update the login info until user restart the app","url":"https://github.com/cline/cline/issues/14751","state":"open","state_reason":null,"labels":["QA","Desktop App"],"created":"2026-10-01T20:49:28Z","updated":"2026-10-06T04:36:22Z","closed":null,"reactions":0,"thumbs_up":0,"comments":2,"severity":"medium","severity_reasons":["class:install/update","class:auth/session"],"topic":"install/update","classes":["install/update","auth/session"],"cosmetic_ui":false,"summary":"I am using a beta version of Cline When user logout from the Cline Desktop app, and navigate to Providers > Configures section, the session still shows that the user is logged in and will not update to logout until user restart the app…","versions":["0.0.40"],"error_snippets":[],"workaround":null,"fixed_in":null,"score":35.4,"first_seen":"2026-10-19T18:38:33Z"},{"id":"cline/cline#15008","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":15008,"title":"[VS Code 4.1.23] OpenAI Compatible selects gpt-4o despite a saved custom model after setup/upgrade","url":"https://github.com/cline/cline/issues/15008","state":"open","state_reason":null,"labels":["VS Code"],"created":"2026-10-10T15:03:02Z","updated":"2026-10-10T15:23:37Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update","self-host/headless","auth/session"],"cosmetic_ui":false,"summary":"VSCode Extension 4.1.23 (saoudrizwan.claude-dev; official v4.1.23 release). After updating an existing Cline installation, the extension returned to onboarding. I configured Bring my own API key → OpenAI Compatible for a self-hosted…","versions":["4.1.23"],"error_snippets":["The model `gpt-4o` does not exist.","2026-10-10T14:37:22.264Z LOG [SessionFactory] Resolved from StateManager: provider=openai, model=undefined, hasApiKey=true","2026-10-10T14:37:26.503Z LOG [VscodeSessionHost] send() completed: text=The model `gpt-4o` does not exist., inputTokens=0","2026-10-10T14:47:15.329Z LOG [SessionFactory] Resolved from StateManager: provider=openai, model=undefined, hasApiKey=true","2026-10-10T14:47:16.874Z LOG [VscodeSessionHost] send() completed: text=The model `gpt-4o` does not exist., inputTokens=0"],"workaround":null,"fixed_in":null,"score":34.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"cline/cline#9248","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":9248,"title":"Cline extension is restarting very often on VS Code with remote SSH connection","url":"https://github.com/cline/cline/issues/9248","state":"closed","state_reason":"not_planned","labels":["P2","Investigation Needed","Needs User Input","VS Code"],"created":"2026-02-11T16:52:17Z","updated":"2026-10-08T09:29:32Z","closed":"2026-09-05T09:08:50Z","reactions":2,"thumbs_up":2,"comments":6,"severity":"medium","severity_reasons":["class:self-host/headless"],"topic":"self-host/headless","classes":["self-host/headless","crash/hang"],"cosmetic_ui":false,"summary":"VSCode Extension While running Cline workflow, Cline extension is restarting very often with VSCode + Remote SSH Connection to Linux VM. Also as per the log analysis of https://github.com/cline/cline/issues/8778 we tried to downgrade to VS…","versions":["3.57.1","1.100.3","1.109.2","10.0.26200"],"error_snippets":["VSCode Extension","While running Cline workflow, Cline extension is restarting very often with VSCode + Remote SSH Connection to Linux VM. Also as per the log analysis of https://github.com/cline/cline/issues/8778 we tried to downgrade to ","3. Cline extension restarts after sometime randomly","VS Code environment details:","Electron: 39.3.0"],"workaround":null,"fixed_in":null,"score":33.7,"first_seen":"2026-10-19T18:38:33Z"},{"id":"kirodotdev/Kiro#11613","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":11613,"title":"Kiro CLI Windows installer fails silently — terminal closes, no log, no error","url":"https://github.com/kirodotdev/Kiro/issues/11613","state":"open","state_reason":null,"labels":["os: windows","cli"],"created":"2026-09-26T01:41:49Z","updated":"2026-10-01T12:00:29Z","closed":null,"reactions":0,"thumbs_up":0,"comments":4,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"This is about the Kiro CLI Windows installer script, not the IDE — routing note for triage. The documented install command is: Under iex the script body executes in the caller's scope. There is no script to exit, so each of the five exit 1…","versions":["1.1.14"],"error_snippets":["iex 'Write-Host \"you should be able to read this\" -ForegroundColor Red; exit 1'","prints the failure, then","$process.ExitCode","Under `iex` the script body executes in the caller's scope. There is no script","to exit, so each of the five `exit 1` calls unwinds the **host** instead —"],"workaround":null,"fixed_in":null,"score":33.1,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aaif-goose/goose#12451","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":12451,"title":"Define private-network access policy for MCP Apps","url":"https://github.com/aaif-goose/goose/issues/12451","state":"open","state_reason":null,"labels":[],"created":"2026-09-22T21:53:50Z","updated":"2026-10-16T11:54:28Z","closed":null,"reactions":0,"thumbs_up":0,"comments":3,"severity":"medium","severity_reasons":["class:mcp/tools"],"topic":"mcp/tools","classes":["mcp/tools","always-on/daemon","network"],"cosmetic_ui":false,"summary":"What problem would this solve? MCP Apps can declare CSP destinations for network connections, resources, frames, and base URIs. Goose currently accepts loopback IPs, localhost, and other IP literals in those declarations while serving app…","versions":[],"error_snippets":["MCP Apps can declare CSP destinations for network connections, resources, frames, and base URIs. Goose currently accepts loopback IPs, `localhost`, and other IP literals in those declarations while serving app content fr","The supported boundary is documented and enforced consistently across connection, resource, frame, and base-URI declarations. Remote or untrusted apps cannot gain surprising access to local services, while legitimate loc","Verification should cover IPv4 and IPv6 loopback, link-local and metadata ranges, private ranges, DNS resolution/rebinding, redirects where relevant, every supported CSP directive, and the approved local-development exce","Localhost preview is an existing MCP Apps use case, so compatibility and migration behavior need an explicit decision before implementation. #11501 contains related destination-policy precedent for a different MCP transp"],"workaround":null,"fixed_in":null,"score":32.9,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aaif-goose/goose#11021","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":11021,"title":"OAuth credentials for streamable HTTP extensions erased by concurrent session starts — endless re-auth loop","url":"https://github.com/aaif-goose/goose/issues/11021","state":"open","state_reason":null,"labels":[],"created":"2026-08-06T19:30:05Z","updated":"2026-09-29T22:22:48Z","closed":null,"reactions":0,"thumbs_up":0,"comments":3,"severity":"critical","severity_reasons":["title:data-loss/security"],"topic":"auth/session","classes":["auth/session","always-on/daemon","mcp/tools"],"cosmetic_ui":false,"summary":"Stored OAuth credentials for a streamable HTTP extension can be erased by Goose itself when multiple sessions start concurrently, producing an endless re-authentication loop: the user completes the vendor's browser auth, and within…","versions":[],"error_snippets":["in sessions.db, created within 54s of each other). All three eagerly connect enabled extensions, including","clear_credentials_on_post_refresh_auth_failure","Stored OAuth credentials for a streamable HTTP extension can be **erased by Goose itself** when multiple sessions start concurrently, producing an endless re-authentication loop: the user completes the vendor's browser a","Timeline from a single machine, **one goosed process running the entire time** (verified via `ps`):","1. `11:40` — user completes Miro browser auth. Keychain secrets blob (service `goose`, account `secrets`) contains a valid `oauth_creds_miro` entry (client_id + token_response). Verified by reading the blob."],"workaround":null,"fixed_in":null,"score":32.9,"first_seen":"2026-10-19T18:38:33Z"},{"id":"continuedev/continue#13076","harness":"continue","harness_name":"Continue","repo":"continuedev/continue","number":13076,"title":"edit_existing_file tool is unusable and crash VSCode.","url":"https://github.com/continuedev/continue/issues/13076","state":"open","state_reason":null,"labels":[],"created":"2026-08-02T23:58:20Z","updated":"2026-10-08T04:12:42Z","closed":null,"reactions":0,"thumbs_up":0,"comments":3,"severity":"high","severity_reasons":["title:breakage"],"topic":"crash/hang","classes":["crash/hang","auth/session","mcp/tools"],"cosmetic_ui":false,"summary":"I've tried finding an answer on the Continue docs site I'm not able to find an open issue that reports the same bug I've seen the troubleshooting guide on the Continue Docs I tried to let my agent use tool editexistingfile but no matter…","versions":["2.0.0","1.0.0"],"error_snippets":["\"arguments\": \"{\\\"filepath\\\":\\\"index.html\\\",\\\"changes\\\":\\\" // ... existing code ...\\\\n let moveSpeed = player.running ? 350 : 500;\\\\n let accel = 1800;\\\\n let friction = player.onGround ? 1500 : 600;\\\\n // ... existing co","SUGGESTED EDIT:","// ... existing code ...","let moveSpeed = player.running ? 350 : 500;","Apply the SUGGESTED EDIT to the ORIGINAL CODE. Output the complete modified file."],"workaround":null,"fixed_in":null,"score":32.9,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aaif-goose/goose#12849","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":12849,"title":"desktop: crash on startup when no theme setting is saved (\"Cannot read properties of undefined (reading 'variant'\")","url":"https://github.com/aaif-goose/goose/issues/12849","state":"open","state_reason":null,"labels":[],"created":"2026-10-12T09:21:16Z","updated":"2026-10-13T21:18:35Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"high","severity_reasons":["title:breakage"],"topic":"crash/hang","classes":["crash/hang"],"cosmetic_ui":false,"summary":"Describe the bug The Goose desktop app shows a \"Honk! An error occurred\" screen with the message: This crashes on startup for users who have never saved a theme preference (i.e., getSetting('theme') returns undefined). The undefined value…","versions":["1.53.0"],"error_snippets":["Cannot read properties of undefined (reading 'variant')","themes[undefined]","The Goose desktop app shows a \"Honk! An error occurred\" screen with the message:","This crashes on startup for users who have never saved a theme preference (i.e., `getSetting('theme')` returns `undefined`). The `undefined` value propagates to `themes[resolvedThemeId].variant` in `ThemeContext.tsx`, wh","3. The app renders the error screen instead of loading."],"workaround":null,"fixed_in":null,"score":32.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"cline/cline#14829","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":14829,"title":"Corrupt @sap/xssec@4.15.0 in bun install silently disabled ~56 @cline/core and ~25 desktop sidecar test files","url":"https://github.com/cline/cline/issues/14829","state":"open","state_reason":null,"labels":[],"created":"2026-10-04T22:59:50Z","updated":"2026-10-04T22:59:53Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"medium","severity_reasons":["class:install/update","class:network"],"topic":"install/update","classes":["install/update","network","always-on/daemon","auth/session"],"cosmetic_ui":false,"summary":"Desktop App + CLI/SDK (affects any test suite whose module graph reaches @cline/llms) A corrupt bun install produced a partially-extracted @sap/xssec@4.15.0 in nodemodules — src/token/ was missing entirely. Because…","versions":["4.15.0","4.9.1"],"error_snippets":["Error: Cannot find module '../token/Token'","produced a partially-extracted","was missing entirely. Because","Cannot find module '../token/Token'","with exit 0 is the real problem. Options: a postinstall smoke check that resolves each"],"workaround":null,"fixed_in":null,"score":32.3,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aaif-goose/goose#13067","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":13067,"title":"FYI: 1 public goose tool-call failure case packaged as a free regression test","url":"https://github.com/aaif-goose/goose/issues/13067","state":"open","state_reason":null,"labels":[],"created":"2026-10-21T23:26:23Z","updated":"2026-10-21T23:26:23Z","closed":null,"reactions":0,"thumbs_up":0,"comments":0,"severity":"high","severity_reasons":["title:breakage"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"I put together a small open dataset of real coding-agent tool-call failures, built from public GitHub issues: https://github.com/faddacake/agent-failure-evals (v0.1). 1 of the 25 cases comes from an issue in this repo. I'm sharing it in…","versions":["4.0"],"error_snippets":["I put together a small open dataset of real coding-agent tool-call failures, built from public GitHub issues: https://github.com/faddacake/agent-failure-evals (v0.1). 1 of the 25 cases comes from an issue in this repo. I"],"workaround":null,"fixed_in":null,"score":32.0,"first_seen":"2026-10-22T09:27:06Z"},{"id":"aaif-goose/goose#12854","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":12854,"title":"load(taskId) returns \"Task not found\" for completed async delegate subagents in Desktop/server sessions (works in standalone CLI)","url":"https://github.com/aaif-goose/goose/issues/12854","state":"closed","state_reason":"completed","labels":[],"created":"2026-10-12T13:14:42Z","updated":"2026-10-15T23:48:45Z","closed":"2026-10-15T23:48:45Z","reactions":1,"thumbs_up":1,"comments":2,"severity":"medium","severity_reasons":["class:self-host/headless"],"topic":"self-host/headless","classes":["self-host/headless","always-on/daemon"],"cosmetic_ui":false,"summary":"Describe the bug In agent sessions running under the goose Desktop app (server-managed backend), results of subagents launched via the delegate tool with async: true cannot be retrieved with load(source: \" \"). The call returns Error: Task…","versions":[],"error_snippets":["You are reproducing a bug for a GitHub issue report. Use ONLY English in everything you write and in every tool call you make. Follow these steps exactly and do nothing else:","1) Call the \"delegate\" tool with: async=true, max_turns=2, and instructions exactly: \"Reply with exactly one line: ENGLISH-REPRO-MARKER-42\"","2) When the subagent completion notification arrives, immediately call the \"load\" tool with source set to that subagent's task ID. Record the exact error message you receive.","3) Call the \"load\" tool with no arguments at all. Record its exact output.","4) Finish with a short English summary listing: the subagent task ID, the exact error from step 2, and the exact output from step 3."],"workaround":null,"fixed_in":null,"score":31.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"cline/cline#14685","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":14685,"title":"Custom provider 'Model Source URL' fetch never sends the API Key -> HTTP 401 on auth-protected /v1/models (Desktop 0.0.38)","url":"https://github.com/cline/cline/issues/14685","state":"open","state_reason":null,"labels":[],"created":"2026-09-30T05:58:06Z","updated":"2026-09-30T16:21:02Z","closed":null,"reactions":0,"thumbs_up":0,"comments":2,"severity":"medium","severity_reasons":["class:auth/session"],"topic":"auth/session","classes":["auth/session","always-on/daemon","self-host/headless","install/update","network"],"cosmetic_ui":false,"summary":"0.0.38 (/Applications/Cline.app, bot.cline.app; hub daemon: code-sidecar --cline-hub-daemon --port 25463 --pathname /hub). Verified against tag desktop-v0.0.38; still present on main as of 2026-09-30. Adding a custom OpenAI-compatible…","versions":["0.0.38"],"error_snippets":["$ curl -s -o /dev/null -w '%{http_code}\\n' http://192.168.99.19:31045/v1/models","{\"error\":{\"code\":\"\",\"message\":\"Invalid token (request id: ...)\",\"type\":\"new_api_error\"}}","export async function fetchModelIdsFromSource(","signal: AbortSignal.timeout(5_000),","throw new Error("],"workaround":null,"fixed_in":null,"score":30.8,"first_seen":"2026-10-19T18:38:33Z"},{"id":"continuedev/continue#13345","harness":"continue","harness_name":"Continue","repo":"continuedev/continue","number":13345,"title":"Optional pre-connect TrustScore signal for MCP servers?","url":"https://github.com/continuedev/continue/issues/13345","state":"open","state_reason":null,"labels":[],"created":"2026-10-05T12:22:02Z","updated":"2026-10-10T23:00:54Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"medium","severity_reasons":["class:self-host/headless","class:mcp/tools"],"topic":"self-host/headless","classes":["self-host/headless","mcp/tools"],"cosmetic_ui":false,"summary":"continue lets users wire in MCP servers as tools, which raises a before-you-connect question: can we show how trustworthy a given MCP server is first? there's a public TrustScore for ~19.7K MCP servers/agents today. an optional check could…","versions":[],"error_snippets":[],"workaround":null,"fixed_in":null,"score":30.8,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aaif-goose/goose#11985","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":11985,"title":"MCP server worked with 1.49 and don't work with 1.50","url":"https://github.com/aaif-goose/goose/issues/11985","state":"closed","state_reason":"completed","labels":[],"created":"2026-09-10T10:44:52Z","updated":"2026-09-21T15:31:55Z","closed":"2026-09-21T15:31:55Z","reactions":1,"thumbs_up":1,"comments":4,"severity":"medium","severity_reasons":["class:self-host/headless","class:mcp/tools"],"topic":"self-host/headless","classes":["self-host/headless","mcp/tools"],"cosmetic_ui":false,"summary":"Describe the bug 💡 Before filing, please check common issues: https://goose-docs.ai/docs/troubleshooting 📦 To help us debug faster, attach your diagnostics zip if possible. 👉 How to capture it…","versions":["1.49","1.50","1.0","1.1"],"error_snippets":["failed to initialize MCP client: uncorrelated error response: expected id 0, error response carried server-error","INFO: 10.0.0.207:54932 - \"POST /macrotracker/mcp HTTP/1.1\" 400 Bad Request","Goose shows me an error:","My MCP server shows an error:","It worked with 1.49 and I expect that it will works with 1.50 also."],"workaround":null,"fixed_in":null,"score":30.6,"first_seen":"2026-10-19T18:38:33Z"},{"id":"cline/cline#14731","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":14731,"title":"plugin-sandbox fails on Windows: Cannot find module 'jiti' silently disables all plugin commands","url":"https://github.com/cline/cline/issues/14731","state":"open","state_reason":null,"labels":["CLI"],"created":"2026-10-01T05:36:22Z","updated":"2026-10-13T14:56:39Z","closed":null,"reactions":0,"thumbs_up":0,"comments":2,"severity":"medium","severity_reasons":["class:mcp/tools"],"topic":"mcp/tools","classes":["mcp/tools","install/update"],"cosmetic_ui":false,"summary":"### Cline Surface 3.0.66 (cline), @cline/core 0.0.88, @cline/cli-windows-x64 3.0.48 I am using a beta version of Cline The plugin-sandbox subprocess fails on every launch and all plugin commands are silently disabled. The failure is…","versions":["3.0.66","0.0.88","3.0.48"],"error_snippets":["plugin-sandbox process exited (code=1, signal=null):","Error: Cannot find module 'jiti'","- C:\\Users\\<user>\\AppData\\Local\\Cline\\[eval]","Error [ERR_MODULE_NOT_FOUND]: Cannot find package '@cline/shared'","C:\\Users\\<user>\\AppData\\Local\\Cline\\[eval]"],"workaround":{"text":"Creating `C:\\Users\\ \\AppData\\Local\\Cline\\node_modules\\` containing `jiti` and `@cline/` makes both errors disappear (verified working).","source":"issue body","url":"https://github.com/cline/cline/issues/14731"},"fixed_in":null,"score":30.1,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aaif-goose/goose#12996","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":12996,"title":"Subagent can never terminate under Code Mode: recipe__final_output is neither a tool nor an execute_typescript callback (regression in 1.54.0)","url":"https://github.com/aaif-goose/goose/issues/12996","state":"closed","state_reason":"completed","labels":[],"created":"2026-10-19T06:14:35Z","updated":"2026-10-23T01:52:16Z","closed":"2026-10-23T01:52:16Z","reactions":0,"thumbs_up":0,"comments":1,"severity":"high","severity_reasons":["title:breakage"],"topic":"install/update","classes":["install/update","crash/hang","mcp/tools"],"cosmetic_ui":false,"summary":"Since 1.54.0, a subagent spawned while Code Mode (codeexecution) is enabled can never terminate. Its system prompt advertises recipefinaloutput and instructs it that calling that tool is mandatory, but under Code Mode the tool is…","versions":["1.54.0","1.55.0"],"error_snippets":["CALL execute_typescript → ReferenceError: Recipe is not defined","CALL execute_typescript → {\"delivered\":null,\"tried\":[\"Recipe__finalOutput -> …\"]} # 13 name variants","SAY \"I have exhausted every available path to that tool.\"","ERROR Action with id \"recipe/final_output\" does not exist","ERROR Action with id \"final_output\" does not exist"],"workaround":null,"fixed_in":null,"score":30.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aaif-goose/goose#13004","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":13004,"title":"Stop and tool timeouts no longer cancel stdio MCP tool calls on 2026-07-28 servers (regression since v1.48.0)","url":"https://github.com/aaif-goose/goose/issues/13004","state":"open","state_reason":null,"labels":[],"created":"2026-10-19T16:44:52Z","updated":"2026-10-21T19:38:07Z","closed":null,"reactions":0,"thumbs_up":0,"comments":0,"severity":"high","severity_reasons":["title:breakage"],"topic":"self-host/headless","classes":["self-host/headless","install/update","mcp/tools","network"],"cosmetic_ui":false,"summary":"Describe the bug Since v1.48.0, goose doesn't send notifications/cancelled when a tool call to a stdio MCP server on protocol 2026-07-28 is stopped (Stop in the desktop, or ACP session/cancel) or hits the extension timeout. goose reports…","versions":["1.48.0","3.4","1.55.0","1.47.0"],"error_snippets":["v1.55.0, 5 s extension timeout","goose: -32603: request timeout after PT5S","edition = \"2021\"","use rmcp::{ErrorData, RoleServer, ServerHandler, ServiceExt};","let line = format!(\"[probe {:>5.1}s] {msg}\\n\", self.started.elapsed().as_secs_f64());"],"workaround":null,"fixed_in":null,"score":30.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"google-gemini/gemini-cli#29518","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":29518,"title":"gemini-3.8-flash is silently forwarded to gemini-3.7-flash for oauth logins","url":"https://github.com/google-gemini/gemini-cli/issues/29518","state":"open","state_reason":null,"labels":["status/need-triage","area/agent"],"created":"2026-09-26T17:20:28Z","updated":"2026-09-29T06:08:26Z","closed":null,"reactions":1,"thumbs_up":1,"comments":1,"severity":"medium","severity_reasons":["class:auth/session"],"topic":"auth/session","classes":["auth/session"],"cosmetic_ui":false,"summary":"Oauth login(Sign in with google) api key login I'd expect to have the model I've selected Client Information Run gemini to enter the interactive CLI, then run the /about command.","versions":["0.61.0","3.7"],"error_snippets":["│ I'm currently processing the core of my identity, specifically focusing on articulating the nature of my model, its generation, and the extent of my knowledge base.","│ This initial phase is about establishing a clear framework for these fundamental descriptors.","│ I'm working to pinpoint the exact details of my knowledge cutoff, aiming for precision regarding the November 2024/January 2025 timeframe. This involves verifying","│ I'm investigating Gemini 3.7 Flash's precise knowledge cutoff, aiming to confirm if it's November 2024 or early 2025 based on available documentation. My","to enter the interactive CLI, then run the"],"workaround":null,"fixed_in":null,"score":29.6,"first_seen":"2026-10-19T18:38:33Z"},{"id":"kirodotdev/Kiro#11522","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":11522,"title":"[Bug] Remote SSH completely broken: wget missing URL → version mismatch → local filesystem fallback","url":"https://github.com/kirodotdev/Kiro/issues/11522","state":"open","state_reason":null,"labels":["os: windows","ssh","theme:ssh-wsl"],"created":"2026-09-22T14:27:37Z","updated":"2026-09-23T16:38:05Z","closed":null,"reactions":0,"thumbs_up":0,"comments":2,"severity":"high","severity_reasons":["title:breakage"],"topic":"self-host/headless","classes":["self-host/headless","install/update"],"cosmetic_ui":false,"summary":"Remote SSH is completely non-functional due to a chain of three cascading bugs. The server binary cannot be downloaded, and even with manual installation, version mismatch causes silent fallback to the local filesystem while the status bar…","versions":["1.1.14","1.95.3","1.96.4"],"error_snippets":["bugs. The server binary cannot be downloaded, and even with manual","In `serverSetup.js` of the `labring.open-remote-ssh-for-trae` extension,","`VERSION_RELEASE_MAP` only contains two entries:"],"workaround":null,"fixed_in":null,"score":29.4,"first_seen":"2026-10-25T09:11:37Z"},{"id":"kirodotdev/Kiro#11502","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":11502,"title":"Kio Down| Auto fallback","url":"https://github.com/kirodotdev/Kiro/issues/11502","state":"open","state_reason":null,"labels":["models","theme:slow-unresponsive","theme:unexpected-error"],"created":"2026-09-21T12:34:47Z","updated":"2026-09-23T16:14:02Z","closed":null,"reactions":0,"thumbs_up":0,"comments":3,"severity":"medium","severity_reasons":["class:crash/hang"],"topic":"crash/hang","classes":["crash/hang"],"cosmetic_ui":false,"summary":"Kiro is down from morning? I have to followup more than 200 times. WTH can't you fallback to available model, if you have availability issues. If none of the model is available, why can't you scale your systems? Kiro is down from morning?…","versions":["1.1.14"],"error_snippets":["WTH can't you fallback to available model, if you have availability issues.","If none of the model is available, why can't you scale your systems?"],"workaround":null,"fixed_in":null,"score":28.6,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aaif-goose/goose#12502","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":12502,"title":"Possible v1.52.0 regression: Goose Desktop loses chat history after restart","url":"https://github.com/aaif-goose/goose/issues/12502","state":"open","state_reason":null,"labels":[],"created":"2026-09-25T01:48:04Z","updated":"2026-10-02T15:00:50Z","closed":null,"reactions":0,"thumbs_up":0,"comments":3,"severity":"high","severity_reasons":["title:breakage"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"A long-running Goose Desktop conversation disappeared from Goose history after upgrading from v1.51.0 to v1.52.0 and subsequently restarting the application. The recurring user-visible behavior was first observed after the v1.51.0 →…","versions":["1.52.0","1.51.0"],"error_snippets":["The conversation is not merely absent from the sidebar. Its canonical session/title record was not found in the active SQLite database or in the preserved database, WAL, SHM, repair, backup, and renderer-state copies exa","The initial loss event was not captured live, so the exact deletion or persistence failure has not been proven.","The following sequence describes the observed failure pattern:","3. Continue using the conversation over an extended period.","8. Search for the previously existing long-running conversation."],"workaround":null,"fixed_in":null,"score":28.6,"first_seen":"2026-10-19T18:38:33Z"},{"id":"cline/cline#14815","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":14815,"title":"Remote helper binary fails with SyntaxError: Invalid character '\\0' (Linux x64, v0.0.43)","url":"https://github.com/cline/cline/issues/14815","state":"closed","state_reason":"completed","labels":[],"created":"2026-10-04T02:36:09Z","updated":"2026-10-13T06:28:08Z","closed":"2026-10-07T22:54:59Z","reactions":1,"thumbs_up":1,"comments":3,"severity":"medium","severity_reasons":["class:self-host/headless"],"topic":"self-host/headless","classes":["self-host/headless","auth/session"],"cosmetic_ui":false,"summary":"Cline Desktop 0.0.43 (darwin/arm64) I am using a beta version of Cline The Linux x64 remote helper shipped in Cline Desktop 0.0.43 crashes immediately on every invocation. Remote environments are completely unusable. The desktop app…","versions":["0.0.43","1.4.2"],"error_snippets":["SyntaxError: Invalid character: '\\0'","Unable to connect to SSH host","Failed to run remote command /home/<user>/.cline/remote/cline-remote-helper-linux-x64-1f54320401bf4e59 (exit 1)","every invocation. Remote environments are completely unusable.","Failed to run remote command /home/ /.cline/remote/cline-remote-helper-linux-x64-1f54320401bf4e59 (exit 1)"],"workaround":null,"fixed_in":null,"score":28.6,"first_seen":"2026-10-21T09:24:21Z"},{"id":"cline/cline#14065","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":14065,"title":"`run_commands` stalls indefinitely before execution when command text contains flagged content - approval await has no timeout, and the 300s auto-proceed never fires","url":"https://github.com/cline/cline/issues/14065","state":"closed","state_reason":"completed","labels":["VS Code"],"created":"2026-09-11T12:42:36Z","updated":"2026-10-07T04:47:56Z","closed":"2026-10-07T04:47:56Z","reactions":0,"thumbs_up":0,"comments":7,"severity":"medium","severity_reasons":["class:network"],"topic":"network","classes":["network","always-on/daemon"],"cosmetic_ui":false,"summary":"VSCode Extension I am using a beta version of Cline Related: #13487 (same symptom family — hang after runcommands, resumed by a user message; CLI, no diagnosis), #13814 + PR #13848 (approval-state recovery, awaitingapproval phase — active…","versions":["4.1.17"],"error_snippets":["{ echo \"T start $(date +%s.%N)\" >> /tmp/repro.log; uptime; echo \"T end $(date +%s.%N)\" >> /tmp/repro.log; }","{ echo \"T start $(date +%s.%N)\" >> /tmp/repro.log; echo 'would run: npm run dev and flask run here'; echo \"T end $(date +%s.%N)\" >> /tmp/repro.log; }","BLOCK=re.compile(r\"(expo start|next dev|npm run dev|uvicorn|flask run)\",re.I)","for t in [\"npx expo start --clear\", \"npm run dev\", \"echo done\"]:","ExecuteCommandToolHandler"],"workaround":null,"fixed_in":null,"score":28.2,"first_seen":"2026-10-24T08:54:58Z"},{"id":"cline/cline#9943","harness":"cline","harness_name":"Cline","repo":"cline/cline","number":9943,"title":"Tips update too frequently causing screen flash, need option to disable tips","url":"https://github.com/cline/cline/issues/9943","state":"closed","state_reason":"not_planned","labels":["stale","JetBrains"],"created":"2026-03-23T21:13:34Z","updated":"2026-10-10T09:43:04Z","closed":"2026-10-10T09:43:04Z","reactions":2,"thumbs_up":2,"comments":3,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update","mcp/tools"],"cosmetic_ui":false,"summary":"JetBrains Plugin The tips text updates way too frequently, causing the screen to flash. This is annoying and distracting. The user expects either a slower update rate or an option to completely turn off tips. Open JetBrains Rider with the…","versions":["3.73.0","2025.3.3"],"error_snippets":["The tips text updates way too frequently, causing the screen to flash. This is annoying and distracting. The user expects either a slower update rate or an option to completely turn off tips.","1. Open JetBrains Rider with the Cline extension installed."],"workaround":null,"fixed_in":null,"score":28.2,"first_seen":"2026-10-25T09:11:37Z"},{"id":"aaif-goose/goose#8638","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":8638,"title":"Goose 1.31.0 panics on startup with existing sessions.db from 1.30.0 (migration v10 orphans sessions)","url":"https://github.com/aaif-goose/goose/issues/8638","state":"closed","state_reason":"completed","labels":[],"created":"2026-04-18T06:46:15Z","updated":"2026-10-16T21:44:01Z","closed":"2026-05-15T16:58:23Z","reactions":1,"thumbs_up":1,"comments":3,"severity":"medium","severity_reasons":["class:install/update","class:crash/hang"],"topic":"install/update","classes":["install/update","crash/hang"],"cosmetic_ui":false,"summary":"After upgrading from Goose 1.30.0 to 1.31.0, starting goose session panics with: The Goose session never starts — the process exits with an error. Schema migration v10 (applied on first run of 1.31.0) introduces a new threads table and…","versions":["1.31.0","1.30.0","0.8.6"],"error_snippets":["thread 'sqlx-sqlite-worker-0' panicked at sqlx-sqlite-0.8.6/src/row.rs:43:46:","Error: no rows returned by a query that expected to return at least one row","sqlite> SELECT COUNT(*) FROM sessions WHERE thread_id IS NULL;","52 -- all existing sessions orphaned","thread_id IS NULL"],"workaround":null,"fixed_in":null,"score":27.9,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aaif-goose/goose#11965","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":11965,"title":"1.50.0 regression: streamable_http MCP extensions fail with \"empty sse stream\" against servers that don't push notifications on the GET channel","url":"https://github.com/aaif-goose/goose/issues/11965","state":"closed","state_reason":"completed","labels":[],"created":"2026-09-09T19:41:32Z","updated":"2026-09-14T20:43:33Z","closed":"2026-09-14T20:43:33Z","reactions":1,"thumbs_up":1,"comments":1,"severity":"high","severity_reasons":["title:breakage"],"topic":"self-host/headless","classes":["self-host/headless","install/update","mcp/tools","network","auth/session"],"cosmetic_ui":false,"summary":"After upgrading from 1.49.0 to 1.50.0, all streamablehttp MCP extensions that worked fine on 1.49.0 fail to connect on 1.50.0, with: and in the CLI log: Bisected to #11827 (\"prefer latest MCP version\"), which changed…","versions":["1.50.0","1.49.0","1.100.0","3.2"],"error_snippets":["Failed to load extension <name>: failed to initialize MCP client: connection closed: discover response","worker quit with fatal: unexpected server response: empty sse stream, when process initialize response","request to establish the server-push SSE channel (per the Streamable HTTP transport). Our server's GET handler (","unexpected server response: empty sse stream","uncorrelated error response"],"workaround":null,"fixed_in":null,"score":27.4,"first_seen":"2026-10-19T18:38:33Z"},{"id":"google-gemini/gemini-cli#29634","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":29634,"title":"bug(cli): gemini.test.tsx fails with FatalUntrustedWorkspaceError when running in non-TTY/headless environments","url":"https://github.com/google-gemini/gemini-cli/issues/29634","state":"open","state_reason":null,"labels":["status/need-triage","area/platform"],"created":"2026-10-05T08:35:11Z","updated":"2026-10-08T13:31:07Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"medium","severity_reasons":["class:self-host/headless"],"topic":"self-host/headless","classes":["self-host/headless","network"],"cosmetic_ui":false,"summary":"In packages/cli/src/gemini.test.tsx, the test 'should NOT load project hooks when workspace is not trusted' fails with FatalUntrustedWorkspaceError whenever tests run in a non-TTY or automated environment (e.g. piped test runners, CI…","versions":[],"error_snippets":["FatalUntrustedWorkspaceError","to enter the interactive CLI, then run the","`FatalUntrustedWorkspaceError` whenever tests run in a non-TTY or automated environment (e.g. piped test runners, CI, background processes, or Windows","terminal environments where `process.stdin.isTTY` is undefined):","FatalUntrustedWorkspaceError: Gemini CLI is not running in a trusted directory. To proceed, either use `--skip-trust`, set the"],"workaround":null,"fixed_in":null,"score":26.8,"first_seen":"2026-10-19T18:38:33Z"},{"id":"kirodotdev/Kiro#11349","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":11349,"title":"## Crash Report #173","url":"https://github.com/kirodotdev/Kiro/issues/11349","state":"open","state_reason":null,"labels":["os: windows","theme:unexpected-error"],"created":"2026-09-12T03:43:29Z","updated":"2026-09-23T16:55:20Z","closed":null,"reactions":0,"thumbs_up":0,"comments":2,"severity":"high","severity_reasons":["title:breakage"],"topic":"crash/hang","classes":["crash/hang","always-on/daemon","self-host/headless","auth/session","network"],"cosmetic_ui":false,"summary":"Date/Time: September 11, 2026, ~04:15 UTC Error Code: 10737 (user-assigned) Task Being Executed: Discussing remaining Blade/Vue fixes after Agent removal What Was Completed Before Crash: Removed Jenssegers\\Agent dependency from 4 files…","versions":["1.0.437"],"error_snippets":["border-edge-top-color","2. JS Syntax Error — `missing } after function body` at feed:1238:1","4. CSS Typos — `border-edge-top-color` should be `border-top-color`"],"workaround":null,"fixed_in":null,"score":26.8,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aaif-goose/goose#12165","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":12165,"title":"Define server identity requirements for externally managed local backends","url":"https://github.com/aaif-goose/goose/issues/12165","state":"open","state_reason":"reopened","labels":[],"created":"2026-09-16T23:24:14Z","updated":"2026-09-24T05:50:40Z","closed":null,"reactions":0,"thumbs_up":0,"comments":2,"severity":"medium","severity_reasons":["class:self-host/headless"],"topic":"self-host/headless","classes":["self-host/headless","always-on/daemon","crash/hang","auth/session"],"cosmetic_ui":false,"summary":"What problem would this solve? Desktop supports externally managed backends through settings and environment configuration, including local development services. Users and backend authors need a clear contract for how the client…","versions":[],"error_snippets":["Desktop supports externally managed backends through settings and environment configuration, including local development services. Users and backend authors need a clear contract for how the client establishes server ide","Decide whether external local services must provide authenticated server identity before receiving the shared secret, or whether an explicitly acknowledged trusted-host development mode remains supported. Define the defa","Consider verified or pinned TLS, another supported server-identity mechanism, and the compatibility implications of an explicit development exception. Apply the decision consistently to startup probes and the subsequent ","This is a transport compatibility/design decision, not a request to implement one approach yet. Acceptance criteria should cover existing local development setups, authenticated external services, and clear failures befo"],"workaround":null,"fixed_in":null,"score":26.8,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aaif-goose/goose#9549","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":9549,"title":"Allow agents to register MCP extensions dynamically at runtime","url":"https://github.com/aaif-goose/goose/issues/9549","state":"open","state_reason":null,"labels":[],"created":"2026-06-01T06:52:46Z","updated":"2026-09-16T19:50:47Z","closed":null,"reactions":0,"thumbs_up":0,"comments":2,"severity":"medium","severity_reasons":["class:mcp/tools"],"topic":"mcp/tools","classes":["mcp/tools","always-on/daemon","self-host/headless","auth/session"],"cosmetic_ui":false,"summary":"Please explain the motivation behind the feature request. Identity/credential brokers (e.g. Warden, and similar projects) sit between agents and external services. At session start, the agent calls the broker's discovery endpoint and gets…","versions":[],"error_snippets":["pub enum ManageExtensionAction {","name: String, // identifier for subsequent Enable/Disable","transport: ExtensionTransport, // StreamableHttp | Stdio | Sse","https://broker.example.com/v1/<ns>/mcp_slack/role/slack-publisher/gateway/",", the extension lives in the session's in-memory extension set (not persisted to config) and is targetable by"],"workaround":null,"fixed_in":null,"score":26.8,"first_seen":"2026-10-26T10:12:02Z"},{"id":"continuedev/continue#11932","harness":"continue","harness_name":"Continue","repo":"continuedev/continue","number":11932,"title":"code-server rendering in Firefox desktop","url":"https://github.com/continuedev/continue/issues/11932","state":"closed","state_reason":"not_planned","labels":["kind:bug","stale"],"created":"2026-03-27T15:13:19Z","updated":"2026-09-26T01:46:33Z","closed":"2026-09-26T01:46:33Z","reactions":1,"thumbs_up":0,"comments":5,"severity":"medium","severity_reasons":["class:self-host/headless"],"topic":"self-host/headless","classes":["self-host/headless","network"],"cosmetic_ui":false,"summary":"I've tried using the \"Ask AI\" feature on the Continue docs site to see if the docs have an answer I'm not able to find a related conversation on GitHub discussions that reports the same bug I'm not able to find an open issue that reports…","versions":["1.3.37","1.2.21","4.112.0","1.1"],"error_snippets":["[HTTP/1.1 404 Not Found 60ms]","[HTTP/1.1 404 Not Found 43ms]","Loading failed for the <script> with source “https://code.cyan.fish/stable-13ca0e47310505f4ab1ac59e891e9a3c5e5eec04/static/node_modules/vsda/rust/web/vsda.js”. code.cyan.fish:1:1","INFO [reconnection-grace-time] Client received grace time from server: 10800000ms (10800s) log.ts:440:13","The web worker extension host is started in a same-origin iframe! webWorkerExtensionHost.ts:121:12"],"workaround":null,"fixed_in":null,"score":26.8,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aaif-goose/goose#12493","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":12493,"title":"UI freezes and \"Install & Restart\" button is unresponsive when updating goose Desktop app on Windows 11","url":"https://github.com/aaif-goose/goose/issues/12493","state":"closed","state_reason":"completed","labels":[],"created":"2026-09-24T18:11:35Z","updated":"2026-10-20T00:04:37Z","closed":"2026-10-20T00:04:37Z","reactions":0,"thumbs_up":0,"comments":4,"severity":"high","severity_reasons":["title:breakage"],"topic":"install/update","classes":["install/update","crash/hang"],"cosmetic_ui":false,"summary":"Description: When attempting to update the goose Desktop application on Windows 11 via the in-app settings (Settings > App > Updates), clicking the Download button works successfully. However, when clicking the subsequent Install & Restart…","versions":[],"error_snippets":["The application completely freezes upon clicking the Install & Restart button. The UI becomes entirely unresponsive, no background installer process is launched, and the app must be forcefully closed via Task Manager or "],"workaround":null,"fixed_in":null,"score":26.2,"first_seen":"2026-10-19T18:38:33Z"},{"id":"google-gemini/gemini-cli#29477","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":29477,"title":"MCP OAuth: RFC 9207 check rejects iss-less callbacks from servers that do not advertise authorization_response_iss_parameter_supported","url":"https://github.com/google-gemini/gemini-cli/issues/29477","state":"closed","state_reason":"completed","labels":["priority/p1","area/security","status/bot-triaged","kind/bug"],"created":"2026-09-24T06:19:04Z","updated":"2026-10-06T22:33:37Z","closed":"2026-10-06T22:33:37Z","reactions":1,"thumbs_up":1,"comments":1,"severity":"critical","severity_reasons":["label:security"],"topic":"self-host/headless","classes":["self-host/headless","auth/session","mcp/tools"],"cosmetic_ui":false,"summary":"Since v0.61.0 (#29117), /mcp auth fails for any MCP server whose authorization server publishes RFC 8414 metadata with an issuer but does not return iss in the authorization response: The same server authenticated fine on v0.56.0. This…","versions":["0.61.0","2.1","2.4"],"error_snippets":["x Failed to authenticate with MCP server '<name>': Missing \"iss\" parameter in authorization response per RFC 9207","if (expectedIssuer) {","// Fail-closed: if an issuer was expected, the response MUST include it","reject(new Error('Missing \"iss\" parameter in authorization response per RFC 9207'));","expectedIssuer"],"workaround":null,"fixed_in":null,"score":26.1,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aws/amazon-q-developer-cli#3923","harness":"amazon-q-cli","harness_name":"Amazon Q Developer CLI","repo":"aws/amazon-q-developer-cli","number":3923,"title":"macOS leaves Kiro CLI HFS+ installer volumes mounted after updates","url":"https://github.com/aws/amazon-q-developer-cli/issues/3923","state":"open","state_reason":null,"labels":[],"created":"2026-10-04T18:28:24Z","updated":"2026-10-04T18:29:02Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"I have run q doctor in the affected terminal session I have run q restart and replicated the issue again macOS 26.5.2 (25F84) Installer / updater image disks are unmounted after an update. I get a growing list of mounted disk images over…","versions":["2.0","1.9","1.4","1.3"],"error_snippets":["blocksize : 512","image-encrypted : false","2026-08-29T21:12:05.538120Z ERROR fig_desktop::install: 102: No update found","2026-09-01T03:21:55.018227Z ERROR fig_install::macos: 252: the update succeeded, but unmounting the dmg failed command=hdiutil: couldn't unmount \"disk4\" - Resource busy","2026-09-02T06:30:41.992268Z ERROR fig_install::macos: 252: the update succeeded, but unmounting the dmg failed command=hdiutil: couldn't unmount \"disk6\" - Resource busy"],"workaround":null,"fixed_in":null,"score":25.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aws/amazon-q-developer-cli#829","harness":"amazon-q-cli","harness_name":"Amazon Q Developer CLI","repo":"aws/amazon-q-developer-cli","number":829,"title":"Login is not persistant","url":"https://github.com/aws/amazon-q-developer-cli/issues/829","state":"closed","state_reason":"completed","labels":["bug"],"created":"2025-03-14T16:44:32Z","updated":"2026-10-22T06:46:34Z","closed":"2026-10-22T06:46:34Z","reactions":2,"thumbs_up":2,"comments":0,"severity":"medium","severity_reasons":["class:auth/session"],"topic":"auth/session","classes":["auth/session","install/update"],"cosmetic_ui":false,"summary":"I have run q doctor in the affected terminal session I have run q restart and replicated the issue again macOS 15.3.2 (24D81) look like the last release introduced a bug in login session not being persisted. i keep getting asked about to…","versions":["1.7.0","64.00","5.9"],"error_snippets":["[environment]","i keep getting asked about to authenticate again every day, it is annoying, i noticed this because everyday the first thing i do is to open the terminal and expect to get autocomplete when typing commands. what i need to"],"workaround":null,"fixed_in":null,"score":24.9,"first_seen":"2026-10-22T09:27:06Z"},{"id":"anomalyco/opencode#49262","harness":"opencode","harness_name":"OpenCode","repo":"anomalyco/opencode","number":49262,"title":"Android/Bionic: Bun 1.4.1 splitting:true causes a.name crash","url":"https://github.com/anomalyco/opencode/issues/49262","state":"closed","state_reason":"completed","labels":[],"created":"2026-09-16T02:37:13Z","updated":"2026-10-01T02:30:04Z","closed":"2026-10-01T02:30:04Z","reactions":2,"thumbs_up":2,"comments":2,"severity":"high","severity_reasons":["title:breakage"],"topic":"crash/hang","classes":["crash/hang"],"cosmetic_ui":false,"summary":"Totally not AI-generated OpenCode 1.18.31 on Android/Bionic crashes before the provider call when compiled with Bun 1.4.1 and splitting:true. Observed failure: TypeError: undefined is not an object (evaluating 'a.name') Affected endpoints…","versions":["1.18.31","1.4.1","0.5.5"],"error_snippets":["TypeError: undefined is not an object (evaluating 'a.name')","Observed failure:","`TypeError: undefined is not an object (evaluating 'a.name')`","Affected endpoints:","So minification and ordinary OpenCode source execution are not required for the bug. The failure tracks specifically with Bun 1.4.1's compiled/embedded `splitting:true` path on Android/Bionic."],"workaround":{"text":"`minify:true, splitting:false`","source":"issue body","url":"https://github.com/anomalyco/opencode/issues/49262"},"fixed_in":null,"score":24.3,"first_seen":"2026-10-19T18:38:33Z"},{"id":"google-gemini/gemini-cli#29405","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":29405,"title":"[Bug Report] Background tasks enter indefinite 'Forever Running' hang due to IPC pipe leaks & missing PID watchdog","url":"https://github.com/google-gemini/gemini-cli/issues/29405","state":"closed","state_reason":"not_planned","labels":["priority/p2","area/agent","status/bot-triaged","kind/bug","status/need-information"],"created":"2026-09-19T04:41:24Z","updated":"2026-10-15T01:45:27Z","closed":"2026-10-15T01:45:27Z","reactions":0,"thumbs_up":0,"comments":3,"severity":"high","severity_reasons":["title:breakage"],"topic":"always-on/daemon","classes":["always-on/daemon","crash/hang","self-host/headless","network"],"cosmetic_ui":false,"summary":"When agent tools execute commands asynchronously in background task mode (runcommand -> managetask), tasks frequently enter an indefinite \"Forever Running\" hang where task status remains permanently stuck on RUNNING. Even after the…","versions":[],"error_snippets":["), tasks frequently enter an indefinite \"Forever Running\" hang where task status remains permanently stuck on",", PowerShell ParserError on line 1, child process exited immediately, but task runner reported","event on all stdio streams, but pipe handles remain unclosed or unflushed, preventing","expect TTY or Ctrl+C. Without non-interactive environment defaults (","When agent tools execute commands asynchronously in background task mode (`run_command` -> `manage_task`), tasks frequently enter an indefinite \"Forever Running\" hang where task status remains permanently stuck on `RUNNI"],"workaround":null,"fixed_in":null,"score":24.2,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aaif-goose/goose#12746","harness":"goose","harness_name":"Goose","repo":"aaif-goose/goose","number":12746,"title":"PreToolUse plugin hooks load but never execute in headless  (main agent)","url":"https://github.com/aaif-goose/goose/issues/12746","state":"closed","state_reason":"completed","labels":[],"created":"2026-10-07T06:12:50Z","updated":"2026-10-21T21:56:08Z","closed":"2026-10-21T21:56:08Z","reactions":0,"thumbs_up":0,"comments":1,"severity":"medium","severity_reasons":["class:self-host/headless","class:mcp/tools"],"topic":"self-host/headless","classes":["self-host/headless","mcp/tools","install/update","crash/hang","auth/session","network"],"cosmetic_ui":false,"summary":"A plugin PreToolUse hook is discovered and loaded (\"Loaded plugin hooks\" appears in the CLI log), but never executes for any tool call made by the main agent in a headless goose run session. The tool runs as if no hook existed. This makes…","versions":["1.50.1"],"error_snippets":["echo \"hook fired\" >> /tmp/hookrepro-fired.log","echo \"denied by policy\" >&2","goose run --text \"Use the Write tool to create leak.txt containing 'leaked'. If denied, quote the denial.\"","{\"level\":\"INFO\",\"fields\":{\"message\":\"Loaded plugin hooks\",\"rule_count\":1,\"events\":\"[\\\"PreToolUse\\\"]\"},\"target\":\"goose::hooks\"}","hook is discovered and loaded (\"Loaded plugin hooks\" appears in the CLI log), but never executes for any tool call made by the **main agent** in a headless"],"workaround":null,"fixed_in":null,"score":24.2,"first_seen":"2026-10-26T10:12:02Z"},{"id":"Aider-AI/aider#5876","harness":"aider","harness_name":"Aider","repo":"Aider-AI/aider","number":5876,"title":"BUG: callable `test_cmd` crashes `Coder.get_platform_info` with `TypeError` on the first message","url":"https://github.com/Aider-AI/aider/issues/5876","state":"open","state_reason":null,"labels":[],"created":"2026-10-18T18:01:02Z","updated":"2026-10-18T18:01:02Z","closed":null,"reactions":0,"thumbs_up":0,"comments":0,"severity":"high","severity_reasons":["title:breakage"],"topic":"crash/hang","classes":["crash/hang","install/update","auth/session"],"cosmetic_ui":false,"summary":"Commands.cmdtest supports a callable test command (testcmd), and Coder.create(..., testcmd=...) accepts one. But Coder.getplatforminfo() builds the system prompt with self.testcmd + \"\\n\", so any coder whose testcmd is a callable raises…","versions":["0.86.0"],"error_snippets":["return \"1 test failed\" # a callable test command returns its error output","1 test failed","/test with the callable test_cmd returns: '1 test failed'","Traceback (most recent call last):","TypeError: unsupported operand type(s) for +: 'function' and 'str'"],"workaround":null,"fixed_in":null,"score":23.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"Aider-AI/aider#5873","harness":"aider","harness_name":"Aider","repo":"Aider-AI/aider","number":5873,"title":"BUG: invalid `--assistant-output-color` is reset to `None`, then every pretty reply crashes in Rich and its edits are not applied","url":"https://github.com/Aider-AI/aider/issues/5873","state":"open","state_reason":null,"labels":[],"created":"2026-10-18T17:56:13Z","updated":"2026-10-18T17:56:13Z","closed":null,"reactions":0,"thumbs_up":0,"comments":0,"severity":"high","severity_reasons":["title:breakage"],"topic":"crash/hang","classes":["crash/hang","install/update","auth/session"],"cosmetic_ui":false,"summary":"InputOutput.validatecolorsettings() handles an invalid assistantoutputcolor by warning \"Disabling this color.\" and setting the attribute to None. The assistant reply is then rendered with rich.markdown.Markdown(..., style=None), and Rich…","versions":["0.86.0"],"error_snippets":["# An invalid assistant_output_color is \"disabled\" to None, then every pretty reply crashes in Rich","# and the reply's edit is not applied.","Applied edit to greeting.txt","Traceback (most recent call last):","File \".../rich/markdown.py\", line 568, in __rich_console__"],"workaround":null,"fixed_in":null,"score":23.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"google-gemini/gemini-cli#29424","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":29424,"title":"bug: Node process hangs on session exit, preventing terminal from returning to shell prompt","url":"https://github.com/google-gemini/gemini-cli/issues/29424","state":"closed","state_reason":"not_planned","labels":["priority/p2","area/agent","status/bot-triaged","kind/bug","status/need-information","effort/large"],"created":"2026-09-20T12:01:43Z","updated":"2026-10-06T01:52:06Z","closed":"2026-10-06T01:52:06Z","reactions":0,"thumbs_up":0,"comments":4,"severity":"high","severity_reasons":["title:breakage"],"topic":"crash/hang","classes":["crash/hang","always-on/daemon","self-host/headless","auth/session","mcp/tools","network"],"cosmetic_ui":false,"summary":"When a Gemini CLI session is finished and displays the termination card (To resume this session: gemini --resume ), the underlying Node.js process hangs and does not exit. This leaves zombie terminal panes in terminal multiplexers like…","versions":["0.60.0","25.3.0"],"error_snippets":["process.exit(0)","When a Gemini CLI session is finished and displays the termination card (`To resume this session: gemini --resume `), the underlying Node.js process hangs and does not exit.","When the session is terminated, the CLI should clean up all of its active event loop handles, close connected MCP server client streams, and cleanly exit to return control back to the shell prompt.","3. Notice the session prints the exit card but hangs.","1. Explicitly close transport streams to connected MCP servers (like `google-workspace` and `chrome-devtools-mcp`)."],"workaround":null,"fixed_in":null,"score":22.8,"first_seen":"2026-10-19T18:38:33Z"},{"id":"continuedev/continue#13335","harness":"continue","harness_name":"Continue","repo":"continuedev/continue","number":13335,"title":"VS Code agent cannot capture terminal output in WSL workspace","url":"https://github.com/continuedev/continue/issues/13335","state":"open","state_reason":null,"labels":[],"created":"2026-10-01T11:34:05Z","updated":"2026-10-03T22:02:56Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"medium","severity_reasons":["class:self-host/headless"],"topic":"self-host/headless","classes":["self-host/headless"],"cosmetic_ui":false,"summary":"Windows with WSL2 (Ubuntu) VS Code opened from a WSL directory using code . Continue VS Code extension installed under both Local and WSL: Ubuntu Continue version: 2.0.0 VS Code version: 1.140.0 When Continue’s agent runs a terminal…","versions":["2.0.0","1.140.0"],"error_snippets":["Continue VS Code extension installed under both Local and WSL: Ubuntu","When Continue’s agent runs a terminal command in the WSL workspace, the command text appears in the VS Code Ubuntu terminal. However, Continue receives neither the command’s output nor its exit status. The tool instead r","This happens even with a harmless command such as pwd. The same command works when I run it myself in the Ubuntu terminal. In a local Windows workspace, Continue runs commands with output visible to the agent, so the fai","Observe that Continue receives the “Output capture is not yet available” message instead of the path or an exit status.","Expected behavior"],"workaround":null,"fixed_in":null,"score":22.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"google-gemini/gemini-cli#29309","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":29309,"title":"bug(core): unbounded _execute recursion on sandbox_expansion_required can loop forever (scheduler.ts)","url":"https://github.com/google-gemini/gemini-cli/issues/29309","state":"closed","state_reason":"not_planned","labels":["priority/p2","area/core","kind/bug","status/manual-triage","status/need-information","effort/medium"],"created":"2026-09-13T13:07:16Z","updated":"2026-09-29T01:53:15Z","closed":"2026-09-29T01:53:15Z","reactions":0,"thumbs_up":0,"comments":5,"severity":"medium","severity_reasons":["class:always-on/daemon"],"topic":"always-on/daemon","classes":["always-on/daemon"],"cosmetic_ui":false,"summary":"packages/core/src/scheduler/scheduler.ts:927-934 recursively calls this.execute() when a tool returns sandboxexpansionrequired, with no counter or depth limit: A malicious or buggy tool returning sandboxexpansionrequired on every call…","versions":[],"error_snippets":["A malicious or buggy tool returning `sandbox_expansion_required` on every call recurses forever (stack growth / infinite loop). The surrounding `catch {}` falls through silently.","Bounded retry: iterative loop with `maxSandboxExpansions` (e.g. 2–3), then fail the call with an error and surface retry count in telemetry. Abort signal should also break the loop.","Fix direction: replace recursion with loop + counter, emit `sandbox_expansion_retry` telemetry, add regression test asserting 3rd consecutive request fails instead of recursing. Related hardening track #29214/#29216."],"workaround":null,"fixed_in":null,"score":22.2,"first_seen":"2026-10-20T09:28:14Z"},{"id":"continuedev/continue#13395","harness":"continue","harness_name":"Continue","repo":"continuedev/continue","number":13395,"title":"Add service-specific Ace media MCP entries to official examples","url":"https://github.com/continuedev/continue/issues/13395","state":"open","state_reason":null,"labels":[],"created":"2026-10-17T19:05:46Z","updated":"2026-10-17T19:05:46Z","closed":null,"reactions":0,"thumbs_up":0,"comments":0,"severity":"medium","severity_reasons":["class:always-on/daemon","class:mcp/tools"],"topic":"always-on/daemon","classes":["always-on/daemon","mcp/tools","self-host/headless","auth/session"],"cosmetic_ui":false,"summary":"Users look for Suno, Seedance, HappyHorse, GPT Image and Nano Banana by service name. Current Continue source removed Hub package/slug loading, so a local MCP file is not an official searchable listing. The existing official Example MCP…","versions":[],"error_snippets":["Users look for Suno, Seedance, HappyHorse, GPT Image and Nano Banana by service name. Current Continue source removed Hub package/slug loading, so a local MCP file is not an official searchable listing. The existing offi","Add five separate service-named sections to that page with one independently loadable remote MCP server each, explicit **by Ace Data Cloud** attribution, token secret references, account access/pricing guidance and links","GPT Image uses the existing OpenAI MCP and must explicitly disclose its broader tool scope and API-token restriction guidance. Async examples must poll the same task ID and never resubmit generation for status. The examp","Readonly initialize/tools-list checks returned HTTP 200 for all five endpoints today; no tools/call or paid media request was made. This issue is the contribution-guide step before the focused docs PR. The no-active-main"],"workaround":null,"fixed_in":null,"score":22.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"google-gemini/gemini-cli#29588","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":29588,"title":"Ink raw mode Ctrl-C unawaited cancellation causes hung interrupts, state rollbacks, and session poisoning (\"Requests ending with a model turn are not         supported\")","url":"https://github.com/google-gemini/gemini-cli/issues/29588","state":"closed","state_reason":"completed","labels":["priority/p1","area/core","status/bot-triaged","kind/bug","effort/medium"],"created":"2026-10-01T02:38:59Z","updated":"2026-10-13T20:02:43Z","closed":"2026-10-13T18:48:55Z","reactions":0,"thumbs_up":0,"comments":3,"severity":"high","severity_reasons":["label:p1"],"topic":"crash/hang","classes":["crash/hang","auth/session"],"cosmetic_ui":false,"summary":"[ACTION REQUIRED] 📎 PLEASE ATTACH THE EXPORTED CHAT HISTORY JSON FILE TO THIS ISSUE IF YOU FEEL COMFORTABLE SHARING IT Root causes: interactiveCli captures 0x03 via Ink keybindings with exitOnCtrlC: false and calls void…","versions":["0.62.0","533.7"],"error_snippets":["[ACTION REQUIRED] 📎 PLEASE ATTACH THE EXPORTED CHAT HISTORY JSON FILE TO THIS ISSUE IF YOU FEEL COMFORTABLE SHARING IT"],"workaround":null,"fixed_in":null,"score":21.5,"first_seen":"2026-10-19T18:38:33Z"},{"id":"google-gemini/gemini-cli#29594","harness":"gemini-cli","harness_name":"Gemini CLI","repo":"google-gemini/gemini-cli","number":29594,"title":"bug(tools): GoogleSearch tool hangs indefinitely in \"Thinking...\" state without timeout","url":"https://github.com/google-gemini/gemini-cli/issues/29594","state":"closed","state_reason":"completed","labels":["priority/p1","area/agent","status/bot-triaged","kind/bug"],"created":"2026-10-01T22:35:14Z","updated":"2026-10-14T05:45:56Z","closed":"2026-10-14T05:45:56Z","reactions":0,"thumbs_up":0,"comments":2,"severity":"high","severity_reasons":["label:p1","title:breakage"],"topic":"crash/hang","classes":["crash/hang","network"],"cosmetic_ui":false,"summary":"When the agent invokes the built-in GoogleSearch tool, the CLI frequently gets stuck in an infinite Thinking... state. The tool execution never times out, never throws an error, and never yields control back to the prompt, keeping the…","versions":["0.61.0"],"error_snippets":["│ GCP Project gemini-enterprise-sec-dx │","ToolExecutionError: Request timed out","to enter the interactive CLI, then run the","immediately aborts the hang, confirming that the terminal event loop is responsive, but the underlying","tool execution promise is left unresolved indefinitely without an"],"workaround":null,"fixed_in":null,"score":21.3,"first_seen":"2026-10-19T18:38:33Z"},{"id":"continuedev/continue#13051","harness":"continue","harness_name":"Continue","repo":"continuedev/continue","number":13051,"title":"Registry entry `dev.continue/docs` points at a docs-site path, not an MCP server","url":"https://github.com/continuedev/continue/issues/13051","state":"open","state_reason":null,"labels":[],"created":"2026-07-28T23:19:26Z","updated":"2026-09-14T08:30:44Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"medium","severity_reasons":["class:self-host/headless","class:mcp/tools"],"topic":"self-host/headless","classes":["self-host/headless","mcp/tools"],"cosmetic_ui":false,"summary":"I probed every remote endpoint advertised in the official MCP registry (10,513 of 10,542 reached). The dev.continue/docs entry came back in the broken column and the reason looks like a straightforward mix-up rather than an outage. The…","versions":["2.0","14.4","12.2","11.5"],"error_snippets":["POST https://docs.continue.dev/mcp -> 405","GET https://docs.continue.dev/mcp -> 404","POST https://docs.continue.dev/api/mcp -> 405","curl -sS -o /dev/null -w '%{http_code}\\n' -X POST https://docs.continue.dev/mcp \\","-H 'content-type: application/json' -H 'accept: application/json, text/event-stream' \\"],"workaround":null,"fixed_in":null,"score":20.8,"first_seen":"2026-10-19T18:38:33Z"},{"id":"kirodotdev/Kiro#11636","harness":"kiro","harness_name":"Kiro","repo":"kirodotdev/Kiro","number":11636,"title":"[Windows] Session snapshot path construction breaks fs_write on absolute paths (ENOENT)","url":"https://github.com/kirodotdev/Kiro/issues/11636","state":"closed","state_reason":"not_planned","labels":["pending-triage"],"created":"2026-09-27T18:23:03Z","updated":"2026-09-30T20:47:21Z","closed":"2026-09-30T20:03:42Z","reactions":1,"thumbs_up":1,"comments":5,"severity":"low","severity_reasons":["low engagement"],"topic":"other","classes":[],"cosmetic_ui":false,"summary":"Windows 11 Pro, version 10.0.26200 (please check Help > About; not captured at report time) fswrite (and any operation that produces a session snapshot) fails with ENOENT whenever the target is an absolute Windows path outside ~/.kiro…","versions":["10.0.26200"],"error_snippets":["Something went wrong: [Error] ENOENT: no such file or directory,","ENOENT whenever the target is an absolute Windows path outside `~/.kiro`.","rather than the extension's `globalStorage` directory, so it may be a","different code path than either of those.","produces two visible symptoms in the same error:"],"workaround":null,"fixed_in":null,"score":20.6,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aws/amazon-q-developer-cli#3915","harness":"amazon-q-cli","harness_name":"Amazon Q Developer CLI","repo":"aws/amazon-q-developer-cli","number":3915,"title":"● failed to retrieve governance settings — MCP and web tools disabled","url":"https://github.com/aws/amazon-q-developer-cli/issues/3915","state":"open","state_reason":null,"labels":[],"created":"2026-09-23T10:49:25Z","updated":"2026-10-09T06:25:18Z","closed":null,"reactions":0,"thumbs_up":0,"comments":1,"severity":"medium","severity_reasons":["class:mcp/tools"],"topic":"mcp/tools","classes":["mcp/tools","self-host/headless"],"cosmetic_ui":false,"summary":"I have run q doctor in the affected terminal session I have run q restart and replicated the issue again Linux 6.6.87.2-microsoft-standard-WSL2 - Ubuntu 20.04.6 LTS kiro cli should start ● failed to retrieve governance settings — MCP and…","versions":["2.23.1","31.15","20.04","20.04.6"],"error_snippets":["[environment]","● failed to retrieve governance settings — MCP and web tools disabled"],"workaround":null,"fixed_in":null,"score":17.8,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aws/amazon-q-developer-cli#3929","harness":"amazon-q-cli","harness_name":"Amazon Q Developer CLI","repo":"aws/amazon-q-developer-cli","number":3929,"title":"远程 SSH 连接 macOS (darwin-arm64)失败","url":"https://github.com/aws/amazon-q-developer-cli/issues/3929","state":"open","state_reason":null,"labels":[],"created":"2026-10-13T07:39:32Z","updated":"2026-10-13T07:39:32Z","closed":null,"reactions":0,"thumbs_up":0,"comments":0,"severity":"medium","severity_reasons":["class:self-host/headless"],"topic":"self-host/headless","classes":["self-host/headless","install/update"],"cosmetic_ui":false,"summary":"I have run q doctor in the affected terminal session I have run q restart and replicated the issue again macOS 26.4.0 (25E246) Windows 客户端升级至 commit a9f293869e317e8f9e195227dffceb57fe52a847 后，远程 SSH 连接 macOS (darwin-arm64) 下载缺失…","versions":["2.30.0","32.00","5.9"],"error_snippets":["[environment]","Windows 客户端升级至 commit a9f293869e317e8f9e195227dffceb57fe52a847 后，远程 SSH 连接 macOS (darwin-arm64) 下载缺失 kiro-reh-darwin-arm64.tar.gz（返回 403 / AccessDenied）。需官方流水线补齐发布"],"workaround":null,"fixed_in":null,"score":17.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aws/amazon-q-developer-cli#3358","harness":"amazon-q-cli","harness_name":"Amazon Q Developer CLI","repo":"aws/amazon-q-developer-cli","number":3358,"title":"bug: q login ignoring identity-provider/region flags","url":"https://github.com/aws/amazon-q-developer-cli/issues/3358","state":"closed","state_reason":"completed","labels":[],"created":"2025-11-04T16:06:28Z","updated":"2026-10-22T02:01:27Z","closed":"2026-10-22T02:01:27Z","reactions":1,"thumbs_up":0,"comments":0,"severity":"medium","severity_reasons":["class:auth/session"],"topic":"auth/session","classes":["auth/session","install/update"],"cosmetic_ui":false,"summary":"I have run q doctor in the affected terminal session I have run q restart and replicated the issue again macOS 15.6.1 If I pass in the --identity-provider or --region flag, I'd expect a browser window or automatic login. If I pass in the…","versions":["1.19.3","36.00","4.1.2"],"error_snippets":["$ q login --license pro --identity-provider https://x-1234567890.awsapps.com/start --region us-east-1","✔ Enter Start URL · https://x-1234567890.awsapps.com/start","✔ Enter Region · us-east-1","[environment]","If I pass in the `--identity-provider` or `--region` flag, I'd expect a browser window or automatic login."],"workaround":null,"fixed_in":null,"score":15.9,"first_seen":"2026-10-22T09:27:06Z"},{"id":"aws/amazon-q-developer-cli#3926","harness":"amazon-q-cli","harness_name":"Amazon Q Developer CLI","repo":"aws/amazon-q-developer-cli","number":3926,"title":"failed to retrieve governance settings — MCP and web tools disabled","url":"https://github.com/aws/amazon-q-developer-cli/issues/3926","state":"open","state_reason":null,"labels":[],"created":"2026-10-10T11:23:45Z","updated":"2026-10-10T11:23:45Z","closed":null,"reactions":0,"thumbs_up":0,"comments":0,"severity":"medium","severity_reasons":["class:mcp/tools"],"topic":"mcp/tools","classes":["mcp/tools","self-host/headless","install/update"],"cosmetic_ui":false,"summary":"I have run q doctor in the affected terminal session I have run q restart and replicated the issue again macOS 15.7.9 (24G830) which kiro-cli kiro-cli 2.29.0 /Users/username/.local/bin/kiro-cli it should properly start with connecting mcp…","versions":["2.29.0","24.00","5.9","17.0.18"],"error_snippets":["[environment]","While the Kiro itself working on Kiro IDE tool but Cli throwing the error.","I open the terminal , type Kiro-cli command and it throws the error .I'm getting the following error:","failed to retrieve governance settings - MCP and web tools disabled","After that, any command I try fails with a \"failure dispatch\" error."],"workaround":null,"fixed_in":null,"score":15.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aws/amazon-q-developer-cli#3914","harness":"amazon-q-cli","harness_name":"Amazon Q Developer CLI","repo":"aws/amazon-q-developer-cli","number":3914,"title":"Auth refresh fails with \"ACP connection closed\" — dispatch failure on every request (CLI 2.23.0, WSL)","url":"https://github.com/aws/amazon-q-developer-cli/issues/3914","state":"closed","state_reason":"completed","labels":[],"created":"2026-09-22T19:30:00Z","updated":"2026-09-23T04:16:52Z","closed":"2026-09-23T04:16:52Z","reactions":0,"thumbs_up":0,"comments":1,"severity":"medium","severity_reasons":["class:self-host/headless","class:auth/session"],"topic":"self-host/headless","classes":["self-host/headless","auth/session","install/update"],"cosmetic_ui":false,"summary":"I have run q doctor in the affected terminal session I have run q restart and replicated the issue again Linux 6.6.87.2-microsoft-standard-WSL2 - Ubuntu 24.04.4 LTS kiro-cli 2.23.0 WSL2 (Ubuntu) on a corporate/enterprise network (MDLZ)…","versions":["2.23.0","1.28.1","7.47","24.04"],"error_snippets":["[environment]","auth.refresh.host_callback_failed: \"Internal error: Failed to verify authentication. Please log in again to continue.\"","acp-peer.refreshToken.error { system: \"acp-peer\", durationMs: ~7000, errorType: \"RequestError\" }","[ActivityLogPublisher] flush failed { error: \"Auth refresh callback failed: Internal error...\" }","Auth refresh callback failed: ACP connection closed"],"workaround":null,"fixed_in":null,"score":13.2,"first_seen":"2026-10-19T18:38:33Z"},{"id":"continuedev/continue#13352","harness":"continue","harness_name":"Continue","repo":"continuedev/continue","number":13352,"title":"Docs scope question: Arcmira remote MCP example","url":"https://github.com/continuedev/continue/issues/13352","state":"open","state_reason":null,"labels":[],"created":"2026-10-07T04:51:53Z","updated":"2026-10-07T04:51:53Z","closed":null,"reactions":0,"thumbs_up":0,"comments":0,"severity":"medium","severity_reasons":["class:self-host/headless","class:mcp/tools"],"topic":"self-host/headless","classes":["self-host/headless","mcp/tools","network"],"cosmetic_ui":false,"summary":"Would a concise Arcmira remote-MCP configuration example fit the existing Example MCP Servers page? I am Arcmira's founder. Following the contribution guide, I wanted to ask about scope before preparing a docs PR. The example would cover…","versions":[],"error_snippets":["Would a concise Arcmira remote-MCP configuration example fit the existing Example MCP Servers page?","The example would cover indexed YouTube transcript research, with account/API-key prerequisites and a link to the canonical setup documentation:","The block passes seven offline assertions using the current upstream MCP-entry schema, including rejection of an invalid transport and a non-string key. This only checks YAML and schema compatibility. I have not tested s","The intent is one short configuration example alongside existing hosted-service examples, with no new dependencies or integration code. Is that within the page's intended scope?"],"workaround":null,"fixed_in":null,"score":11.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aws/amazon-q-developer-cli#3921","harness":"amazon-q-cli","harness_name":"Amazon Q Developer CLI","repo":"aws/amazon-q-developer-cli","number":3921,"title":"Startup stalls ~2s before first frame: shim walks all of /proc, cost scales with machine thread count","url":"https://github.com/aws/amazon-q-developer-cli/issues/3921","state":"open","state_reason":null,"labels":[],"created":"2026-10-02T11:05:04Z","updated":"2026-10-02T11:11:11Z","closed":null,"reactions":0,"thumbs_up":0,"comments":0,"severity":"low","severity_reasons":["low engagement"],"topic":"install/update","classes":["install/update","network"],"cosmetic_ui":false,"summary":"I have run q doctor in the affected terminal session I have run q restart and replicated the issue again Linux Red Hat Enterprise Linux 8.10 (Ootpa) kiro-cli chat launch doesn't take super long kiro-cli chat launch takes super long and is…","versions":["2.21.0","2.21.4","2.23.0","2.25.0"],"error_snippets":["Linux Red Hat Enterprise Linux 8.10 (Ootpa)","During that stall the shim walks the entire `/proc` tree. The cost is","43913 open (38902 fail ENOENT)","open(\"/proc/ /task/ /task\", O_RDONLY) = -1 ENOENT","That path is one level deeper than exists; `task` is not nested under a thread."],"workaround":null,"fixed_in":null,"score":7.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"Aider-AI/aider#5776","harness":"aider","harness_name":"Aider","repo":"Aider-AI/aider","number":5776,"title":"segfault while listing models on 0.86.2","url":"https://github.com/Aider-AI/aider/issues/5776","state":"open","state_reason":null,"labels":[],"created":"2026-09-26T16:22:13Z","updated":"2026-09-26T16:22:13Z","closed":null,"reactions":0,"thumbs_up":0,"comments":0,"severity":"medium","severity_reasons":["class:crash/hang"],"topic":"crash/hang","classes":["crash/hang","install/update"],"cosmetic_ui":false,"summary":"aider 0.86.2 installed freshly via pip in a venv","versions":["0.86.2"],"error_snippets":["Segmentation fault"],"workaround":null,"fixed_in":null,"score":1.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"aws/amazon-q-developer-cli#3911","harness":"amazon-q-cli","harness_name":"Amazon Q Developer CLI","repo":"aws/amazon-q-developer-cli","number":3911,"title":"cloudConfig.getManifest and ActivityLogPublisher return 401 Unable to resolve user ID after upgrade to 2.21.4","url":"https://github.com/aws/amazon-q-developer-cli/issues/3911","state":"open","state_reason":null,"labels":[],"created":"2026-09-16T08:28:11Z","updated":"2026-09-16T08:28:11Z","closed":null,"reactions":0,"thumbs_up":0,"comments":0,"severity":"medium","severity_reasons":["class:install/update","class:auth/session"],"topic":"install/update","classes":["install/update","auth/session","self-host/headless"],"cosmetic_ui":false,"summary":"I have run q doctor in the affected terminal session I have run q restart and replicated the issue again Linux 6.18.33.2-microsoft-standard-WSL2 - Debian GNU/Linux 13 (trixie) Description: After upgrading kiro-cli to 2.21.4, every session…","versions":["2.21.4"],"error_snippets":["[environment]","After upgrading kiro-cli to 2.21.4, every session logs a 401 Unauthorized from two backend calls, even when the session is otherwise authenticated","errorType: UnauthorizedException, statusCode: 401, message: \"Unable to resolve user ID\"","2. ActivityLogPublisher -> https://runtime.us-east-1.kiro.dev/agents/activity","Failed to publish to https://runtime.us-east-1.kiro.dev/agents/activity: 401 — {\"message\":\"Unauthorized\"}"],"workaround":null,"fixed_in":null,"score":0.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"Aider-AI/aider#5719","harness":"aider","harness_name":"Aider","repo":"Aider-AI/aider","number":5719,"title":"Hexum hosted MCP for smaller coding-agent context?","url":"https://github.com/Aider-AI/aider/issues/5719","state":"open","state_reason":null,"labels":[],"created":"2026-09-16T17:44:58Z","updated":"2026-09-16T17:44:58Z","closed":null,"reactions":0,"thumbs_up":0,"comments":0,"severity":"medium","severity_reasons":["class:mcp/tools"],"topic":"mcp/tools","classes":["mcp/tools"],"cosmetic_ui":false,"summary":"Hi Aider maintainers — Hexum offers a hosted MCP server at https://hexum.dev/mcp that shrinks coding-agent turn context, with optional architecture.hexum checks. Would you consider listing/linking it for users, or would a short integration…","versions":[],"error_snippets":[],"workaround":null,"fixed_in":null,"score":0.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"continuedev/continue#13273","harness":"continue","harness_name":"Continue","repo":"continuedev/continue","number":13273,"title":"CLI: runTerminalCommand \"should handle non-existent commands\" intermittently hangs to vitest's 30s timeout on windows-latest","url":"https://github.com/continuedev/continue/issues/13273","state":"open","state_reason":null,"labels":[],"created":"2026-09-14T00:17:40Z","updated":"2026-09-14T00:17:40Z","closed":null,"reactions":0,"thumbs_up":0,"comments":0,"severity":"high","severity_reasons":["title:breakage"],"topic":"crash/hang","classes":["crash/hang","network","self-host/headless"],"cosmetic_ui":false,"summary":"I've tried finding an answer on the Continue docs site I'm not able to find an open issue that reports the same bug I've seen the troubleshooting guide on the Continue Docs src/tools/runTerminalCommand.test.ts > runTerminalCommandTool >…","versions":[],"error_snippets":["extensions/cli, vitest, windows-latest GitHub runner","❯ src/tools/runTerminalCommand.test.ts (8 tests | 1 failed) 34086ms","✓ basic platform-specific terminal execution > should execute a simple echo command 460ms","✓ basic platform-specific terminal execution > should get current directory 661ms","✓ basic platform-specific terminal execution > should list directory contents 427ms"],"workaround":null,"fixed_in":null,"score":0.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"continuedev/continue#13295","harness":"continue","harness_name":"Continue","repo":"continuedev/continue","number":13295,"title":"Update Recommended Models","url":"https://github.com/continuedev/continue/issues/13295","state":"open","state_reason":null,"labels":["area:docs"],"created":"2026-09-19T00:49:04Z","updated":"2026-09-19T00:49:04Z","closed":null,"reactions":0,"thumbs_up":0,"comments":0,"severity":"medium","severity_reasons":["class:install/update"],"topic":"install/update","classes":["install/update"],"cosmetic_ui":false,"summary":"Incorrect or outdated information https://docs.continue.dev/customize/models It's already very old information.","versions":[],"error_snippets":[],"workaround":null,"fixed_in":null,"score":0.0,"first_seen":"2026-10-19T18:38:33Z"},{"id":"continuedev/continue#13303","harness":"continue","harness_name":"Continue","repo":"continuedev/continue","number":13303,"title":"for mcp's with 50-100 tools apply a state to all \"ask first / automatica / excluded\"","url":"https://github.com/continuedev/continue/issues/13303","state":"open","state_reason":null,"labels":[],"created":"2026-09-21T08:06:22Z","updated":"2026-09-21T08:06:22Z","closed":null,"reactions":0,"thumbs_up":0,"comments":0,"severity":"medium","severity_reasons":["class:mcp/tools"],"topic":"mcp/tools","classes":["mcp/tools"],"cosmetic_ui":false,"summary":"I've tried finding an answer on the Continue docs site I'm not able to find an open issue that reports the same bug I've seen the troubleshooting guide on the Continue Docs","versions":[],"error_snippets":["Super annoying to have to set 100 tools manually. Easier would be"],"workaround":null,"fixed_in":null,"score":0.0,"first_seen":"2026-10-19T18:38:33Z"}]}