WORKING: All Microsoft Issues and PRs (refresh)#74163
WORKING: All Microsoft Issues and PRs (refresh)#74163BradGroux wants to merge 6 commits intoopenclaw:mainfrom
Conversation
Greptile SummaryThis PR adds
Confidence Score: 4/5Doc-only change is safe to merge after fixing the stale "PR body" instruction; no code is affected. One P1 finding (wrong instruction text that will mislead maintainers) and one P2 finding (miscategorized macOS issue). No code changes, no logic paths affected. Score is at the P1 ceiling of 4. MICROSOFT_TRACKER.md — the "How to use" instruction on line 13 needs to be corrected from "PR body" to "this file", and issue #64443 on line 142 should be reviewed for correct categorization. Prompt To Fix All With AIThis is a comment left during a code review.
Path: MICROSOFT_TRACKER.md
Line: 13
Comment:
**Stale instruction: references "this PR body" instead of the file**
This tracker now lives as a repository file, not a PR body. The instruction to "edit this PR body" will mislead maintainers who look here for update guidance — there is no associated PR body to edit; changes need to go through a commit to `MICROSOFT_TRACKER.md`.
```suggestion
> - Mark items resolved by editing this file and changing `[ ]` to `[x]`
```
How can I resolve this? If you propose a fix, please make it concise.
---
This is a comment left during a code review.
Path: MICROSOFT_TRACKER.md
Line: 142
Comment:
**macOS item miscategorized under Windows Platform**
Issue #64443 title is "OpenClaw chat interface causes very high **WindowServer** CPU on Intel Retina iMac." `WindowServer` is the macOS display compositor — it does not exist on Windows. This item appears to have been pulled in via a keyword match on "Window" rather than a genuine Windows-platform signal, and belongs under a macOS/UI category or should be removed from this tracker entirely.
How can I resolve this? If you propose a fix, please make it concise.Reviews (1): Last reviewed commit: "docs: refresh Microsoft ecosystem tracke..." | Re-trigger Greptile |
| > - Items marked _(stale)_ have been flagged by the stale bot due to inactivity | ||
| > | ||
| > **Last updated:** 2026-04-29 (post-purge audit: refreshed from currently open GitHub issues/PRs and rebuilt from PR #49126 format) | ||
|
|
There was a problem hiding this comment.
Stale instruction: references "this PR body" instead of the file
This tracker now lives as a repository file, not a PR body. The instruction to "edit this PR body" will mislead maintainers who look here for update guidance — there is no associated PR body to edit; changes need to go through a commit to MICROSOFT_TRACKER.md.
| > - Mark items resolved by editing this file and changing `[ ]` to `[x]` |
Prompt To Fix With AI
This is a comment left during a code review.
Path: MICROSOFT_TRACKER.md
Line: 13
Comment:
**Stale instruction: references "this PR body" instead of the file**
This tracker now lives as a repository file, not a PR body. The instruction to "edit this PR body" will mislead maintainers who look here for update guidance — there is no associated PR body to edit; changes need to go through a commit to `MICROSOFT_TRACKER.md`.
```suggestion
> - Mark items resolved by editing this file and changing `[ ]` to `[x]`
```
How can I resolve this? If you propose a fix, please make it concise.| | [ ] | P2 | #63491 | [Bug]: Windows Scheduled Task gateway restart/health becomes inconsistent after ready | | | | ||
| | [ ] | P2 | #60713 | Windows: resolvePreferredOpenClawTmpDir uses C:\tmp instead of proper temp directory | | | | ||
| | [ ] | P2 | #59709 | Cron scheduled task delivery fails (v2026.3.31) | | | | ||
| | [ ] | P2 | #59281 | [Bug]: Windows plugin TS source-loading via jiti is pathologically slow in real production call sites | `bug` `bug:behavior` | | |
There was a problem hiding this comment.
macOS item miscategorized under Windows Platform
Issue #64443 title is "OpenClaw chat interface causes very high WindowServer CPU on Intel Retina iMac." WindowServer is the macOS display compositor — it does not exist on Windows. This item appears to have been pulled in via a keyword match on "Window" rather than a genuine Windows-platform signal, and belongs under a macOS/UI category or should be removed from this tracker entirely.
Prompt To Fix With AI
This is a comment left during a code review.
Path: MICROSOFT_TRACKER.md
Line: 142
Comment:
**macOS item miscategorized under Windows Platform**
Issue #64443 title is "OpenClaw chat interface causes very high **WindowServer** CPU on Intel Retina iMac." `WindowServer` is the macOS display compositor — it does not exist on Windows. This item appears to have been pulled in via a keyword match on "Window" rather than a genuine Windows-platform signal, and belongs under a macOS/UI category or should be removed from this tracker entirely.
How can I resolve this? If you propose a fix, please make it concise.There was a problem hiding this comment.
Pull request overview
Reintroduces and refreshes a maintainer-facing Microsoft ecosystem issue/PR tracker (Teams, Windows, WSL, Azure, M365/SharePoint) after the purge, keeping the prior checklist/table structure and updating counts to currently-open items.
Changes:
- Adds
MICROSOFT_TRACKER.mdwith refreshed open issue/PR inventory and summary counts by Microsoft-related area. - Includes appendix views for P0 blockers, high-priority regressions, and stale items.
- Adds audit notes describing the regenerated source set and inclusion criteria.
| > | ||
| > **How to use:** | ||
| > | ||
| > - Mark items resolved by editing this PR body and changing `[ ]` to `[x]` |
| ### Bugs / Crashes | ||
|
|
||
| | Resolved? | Priority | # | Title | Labels | Assignee | | ||
| | --------- | -------- | - | ----- | ------ | -------- | |
| ## Appendix: High-Priority Bugs / Regressions | ||
|
|
||
| | Category | Type | # | Title | | ||
| | -------- | ---- | - | ----- | |
|
Codex review: found issues before merge. Summary Reproducibility: yes. for the review findings: PR-head source shows the stale committed counts, hard-coded closed-count reset, stale edit instruction, and substring matcher. No product runtime reproduction applies to this maintainer tracker PR. Real behavior proof Next step before merge Security Review findings
Review detailsBest possible solution: Keep the draft open, choose the intended source of truth, then refresh the committed tracker and generator so they preserve resolved rows, current counts, valid categories, and the owner workflow before merge. Do we have a high-confidence way to reproduce the issue? Yes for the review findings: PR-head source shows the stale committed counts, hard-coded closed-count reset, stale edit instruction, and substring matcher. No product runtime reproduction applies to this maintainer tracker PR. Is this the best way to solve the issue? No as currently proposed; the tracker-generator direction is plausible, but the branch should not merge until the committed file and generator match the active maintainer workflow and current tracker state. Full review comments:
Overall correctness: patch is incorrect Acceptance criteria:
What I checked:
Likely related people:
Remaining risk / open question:
Codex review notes: model gpt-5.5, reasoning high; reviewed against 9c20f4352024. |
|
Microsoft tracker audit complete (2026-05-04 local time). I updated the PR body only; no code or repo files were changed. Summary:
New open items added: Resolved since last audit: Removed from tracker as false positives after review: Notes:
|
|
Tracker update: #77907/#77785 marked as maintainer-prepped/rebased with fresh CI pending. Counts unchanged until the PR merges and the issue closes. PR comment: #77907 (comment) |
|
Tracker update: #78081/#77674 marked as maintainer-prepped/rebased with changelog added, maintainer proof override posted, and fresh CI pending. Counts unchanged until merge/closure. PR comment: #78081 (comment) |
|
Tracker refresh: #77907/#77785 is now maintainer-refreshed at prepared head |
|
Tracker update: #77907 merged at head |
|
Tracker update after #78081 merge:
Next low-hanging candidates while the queue is warm: #78159, #77993/#78006/#78171, #77958/#77971, then the older XS Windows PRs that need rebase/conflict cleanup (#60678/#62910/#48320). |
|
Refreshed stale/conflicting prepared PRs after main moved.\n\n- #60678 rebased/pushed from |
|
Low-hanging fruit pass complete (2026-05-08): Merged/resolved:
Prepared / waiting:
Counts updated:
|
|
Tracker updated
Counts updated:
|
|
Follow-up tracker correction: added #62742 to the Azure PR list, it was missing from the tracker. Current #62742 state is open, prepared earlier at Adjusted counts after adding #62742:
|
|
Tracker update:
Counts unchanged for now because both items are prepared but not merged/closed yet. |
|
Tracker update:
No counts changed because neither PR has merged/closed yet. |
|
Tracker update:
No counts changed because none merged or closed in this pass. |
|
Maintainer batch update for the Microsoft tracker: prepared five PRs and updated the tracker body with the current heads/status. Prepared and waiting on fresh CI/ClawSweeper:
No resolved-count change yet; these remain open until fresh CI and maintainer review clear them for merge. |
|
Tracker update after maintainer merge:
No other tracked Microsoft PRs met the merge criteria in this pass. |
|
Maintainer update for 2026-05-09: Closed/resolved tracker items:
Prepared for follow-up CI/review:
Counts updated:
|
|
Tracker update for the current Microsoft pass: Resolved/closed since the post-purge tracker refresh:
Counts updated:
Checked #79694 as a possible next merge candidate. It is small and mergeable with current GitHub checks green, but the durable ClawSweeper review still asks for live Windows behavior proof, so I did not merge it without platform-real evidence. |
Microsoft Ecosystem Issues & PRs Tracker
Summary
1. MS Teams Channel Plugin — Issues
Bugs / Crashes
Feature Requests
2. MS Teams Channel Plugin — PRs
3. Windows Platform — Issues
Bugs / Crashes
node-llama-cppWindows binaries crash (0xC0000005) on Intel Alder Lake-N (N95) — qmd LLM half unusablebugbug:crashbugregressionbugregressionbugregressionbugregressionbugregressionbugregressionbugbug:crashbugbug:crashchannels.slack.start-accountphase blocks event loop 5+ minutes while a model_call is in flight (Windows, 2026.5.4)bugbugbug:behaviorbugbugbugbugbug:behaviorbugbug:behaviorbugbug:behaviorbugbug:behaviorbugbug:behaviorpnpm check:docswork in native PowerShellopenclaw updatecommand fails with EBUSY error on Windowsbugbug:behaviorFeature Requests
enhancementcronfield to Agent configuration for Agent-owned scheduled tasksenhancementenhancementenhancementhelp wanted4. Windows Platform — PRs
5. WSL (Windows Subsystem for Linux) — Issues
Bugs / Crashes
bugbugbug:crashbugregressionbugregressionbugFeature Requests
enhancement6. WSL (Windows Subsystem for Linux) — PRs
7. Azure (Provider / Infrastructure) — Issues
Bugs / Crashes
Feature Requests
enhancement8. Azure (Provider / Infrastructure) — PRs
9. Microsoft 365 / SharePoint — Issues
Bugs / Crashes
No currently open items found.
Feature Requests
No currently open items found.
10. Microsoft 365 / SharePoint — PRs
No currently open items found.
Appendix: P0 Blockers (Start Here)
Appendix: High-Priority Bugs / Regressions
node-llama-cppWindows binaries crash (0xC0000005) on Intel Alder Lake-N (N95) — qmd LLM half unusableAppendix: Stale Items (Consider Closing)
No matching items found.
Audit Notes
openclaw/openclaw; closed counts are intentionally reset to0for this refreshed tracker.msteams, Microsoft Teams, Windows, WSL, Azure, Entra/AAD, MSAL, managed identity, DefaultAzureCredential, Microsoft Graph, SharePoint, OneDrive, and Microsoft 365.channel: msteams, because those can still affect the Microsoft surface area.node scripts/generate-microsoft-tracker.mjsso the tracker and PR body can be refreshed after future triage passes.