feat: pulse CI failure pattern detection — close self-improvement gap from GH#2973#2976
Conversation
…kflow bugs The pulse processed PRs individually without correlating CI failures across them. When the same check failed on all PRs (e.g., opencode-agent regex false positive, review-bot-gate concurrency cancellation), the pulse either dispatched workers to fix individual PRs (useless — the bug is in the workflow, not the PR code) or ignored the failures entirely. Added a 'CI failure pattern detection' section to pulse.md Step 3 that: - Correlates failing check names across all open PRs in each repo - Identifies checks failing on 3+ PRs as likely systemic (workflow bugs) - Files a single issue for systemic failures instead of dispatching N workers - Continues dispatching workers for per-PR failures as before This closes the self-improvement gap that allowed GH#2973 to go undetected. Closes #2975
|
Warning You have reached your daily quota limit. Please wait up to 24 hours and I will start processing your requests again! |
🔍 Code Quality Report�[0;35m[MONITOR]�[0m Code Review Monitoring Report �[0;34m[INFO]�[0m Latest Quality Status: �[0;34m[INFO]�[0m Recent monitoring activity: 📈 Current Quality Metrics
Generated on: Fri Mar 6 04:53:33 UTC 2026 Generated by AI DevOps Framework Code Review Monitoring |
|
Caution Review failedPull request was closed or merged during review WalkthroughDocumentation is added to pulse.md outlining CI failure pattern detection: identifying systemic failures when the same check fails across 3+ open PRs, distinguishing them from per-PR failures, and prescribing workflow actions (halt dispatch, file issues, apply labels). Changes
Estimated code review effort🎯 1 (Trivial) | ⏱️ ~3 minutes Possibly related PRs
Poem
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
|



Summary
Adds CI failure pattern detection to the pulse supervisor (Step 3 of
pulse.md), closing the self-improvement gap that allowed GH#2973 to go undetected.Problem
The pulse processed PRs individually without correlating CI failures across them. When the same workflow check failed on all 10 open PRs (opencode-agent regex false positive + review-bot-gate concurrency cancellation), the pulse either:
mergeStateStatus: UNSTABLEwasn't actionableThe self-improvement signals in AGENTS.md already mentioned "Multiple PRs fail CI with the same error pattern" as a signal, but the pulse had no implementation to actually detect this.
Fix
Added a new subsection "CI failure pattern detection" to pulse.md Step 3 that instructs the pulse agent to:
This is guidance for the LLM agent (not a deterministic script), consistent with the "Intelligence Over Determinism" principle — the pulse agent reads check names, correlates across PRs, and uses judgment to distinguish systemic from per-PR failures.
Related
Closes #2975
Summary by CodeRabbit
Documentation