Repository navigation
feat: implement issue #401 — [Fleet Monitor] petry-projects/.github-private — issue-triage-runner.yml - #476
Conversation
…rivate — issue-triage-runner.yml
|
Warning Review limit reached
More reviews will be available in 27 minutes and 45 seconds. Learn how PR review limits work. Your organization has run out of usage credits. Purchase more in the billing tab. ⌛ How to resolve this issue?After more reviews become available, a review can be triggered using the We recommend that you space out your commits to avoid hitting the rate limit. 🚦 How do rate limits work?CodeRabbit enforces hourly rate limits for each developer per organization. Our paid plans include higher PR review limits than trial, open-source, and free plans. In all cases, reviews become available again over time. During sustained high-volume PR review activity, CodeRabbit may temporarily slow when the next review becomes available. Please see our Fair Usage Limits Policy for further information. ℹ️ Review info⚙️ Run configurationConfiguration used: Organization UI Review profile: ASSERTIVE Plan: Pro Run ID: 📒 Files selected for processing (2)
✨ 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 |
There was a problem hiding this comment.
Code Review
This pull request adds a new integration test to verify that the safe-output command successfully validates allowed labels and comments. The reviewer recommended mocking the gh CLI tool during this test to prevent external network calls or command-not-found errors, making the test hermetic and avoiding the need to mask failures with || true.
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 4e57c9f235
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
Dev-Lead — rate-limited (intent: fix-reviews)PR: #476 |
Dev-Lead — fix-reviews (applied)Changes committed and pushed. |
Dev-Lead — waiting on PR blockers (intent: review-changes)PR: #476 |
|
Note @don-petry I reviewed this PR and no code changes were needed, but it still has blocking checks or reviews (failing or cancelled checks, or changes-requested reviews), so I cannot mark it done yet. I'll re-check automatically. |
Dev-Lead — review-changes (no-changes)No changes were needed for this PR. |
donpetry-bot
left a comment
There was a problem hiding this comment.
Automated review — APPROVED ✓
Risk: LOW
Reviewed commit: 62935572bb13afc044d8e4c735b5d93bb53ff5ee
Review mode: triage-approved (single reviewer)
Summary
Small, well-scoped bash bug fix in scripts/aw.sh that aligns the frontmatter lookup paths with the actual workflow schema and removes a now-unreachable code path. Adds a hermetic safe-output validation test. Triage tier's low-risk classification is confirmed.
Linked issue analysis
Closes #401 — [Fleet Monitor] issue-triage-runner.yml at 100% failure rate (11/11 runs). The fix targets the most likely root cause: scripts/aw.sh was reading allowed-labels and max-labels from the top level of the workflow frontmatter, but .github/workflows/issue-triage.md actually defines them under safe-outputs.add-labels.allowed and safe-outputs.add-labels.max. This schema mismatch would consistently break the triage runner. The PR updates both python3 extractors to the correct nested path.
Findings
- Schema fix (correct):
scripts/aw.shnow readsfm.get("safe-outputs", {}).get("add-labels", {}).get("allowed", [])and.get("max", 3), matchingissue-triage.mdexactly. - Dead-code removal (safe): The
issue_action-based "skip on reopen if any current label is in allowed set" branch is removed along with the local var and fixture lookup. The corresponding workflow only listens forissues: types: [opened], so thereopenedbranch was unreachable in practice — removal does not change observable behavior. - New test (good):
tests/aw/issue-triage/test_aw_run.shadds a hermetic safe-output happy-path test that mocks theghCLI onPATH(exit 0), directly addressing gemini-code-assist's earlier note about avoiding network calls /|| truemasking. - No security concerns: No changes to auth, secrets, env handling, or shell-injection surface. Existing quoted-heredoc + env-var pattern for user-controlled issue data is preserved.
CI status
All required checks passing: ShellCheck, shellcheck (aw-*.sh), bats, gh-aw-compile, validate-agent-profiles, CodeQL (actions + python), Secret scan (gitleaks), Agent Security Scan, Compile agentic workflows, unit-tests, AgentShield, SonarCloud (Quality Gate passed), Validate AW specs/docs/workflows, dev-lead dispatch, PR Review Agent. CodeRabbit was rate-limited (no findings); Copilot errored (transient); Codex left no suggestions; gemini's suggestion was applied.
Reviewed automatically by the PR-review agent (single-reviewer mode: opus 4.7). Reply if you need a human review.
Dev-Lead — review-changes (no-changes)No changes were needed for this PR. |
|
donpetry-bot
left a comment
There was a problem hiding this comment.
Automated review — APPROVED ✓
Risk: MEDIUM
Reviewed commit: 6b2385509af78b3f5b721b4538ae70a9c68e1ada
Cascade: triage → deep (triage: haiku 4.5 → deep: sonnet 4.6 + duck: o4-mini → audit: opus 4.7)
Summary
This PR correctly fixes a schema mismatch: aw.sh was reading allowed-labels and max-labels as flat top-level frontmatter keys that never existed in issue-triage.md, causing 100% failure. The reopen-prevention logic removed was genuinely dead code because the workflow only fires on issues.types=[opened], never reopened. All CI gates pass and the triage signals are fully explained by analysis.
Findings
- INFO: Schema fix is correct: aw.sh now reads fm.get('safe-outputs',{}).get('add-labels',{}).get('allowed',[]) and .get('max',3), matching the actual nested structure in issue-triage.md. The previous flat keys allowed-labels and max-labels never existed, causing every label to be rejected and a 100% failure rate.
- INFO: Reopen-prevention logic (skip if has_triage_label && issue_action == reopened) was dead code. The issue-triage.md workflow fires only on issues.types=[opened], so issue_action could never be 'reopened' at runtime. Removal is safe and does not change observable behavior.
- INFO: New Test 6 in test_aw_run.sh adds a hermetic safe-output happy-path test using a mock gh CLI (exit 0), directly validating that valid labels+comment produce 'validation passed'. Addresses the gemini-code-assist recommendation to avoid network calls and || true masking.
- INFO: All required checks pass: ShellCheck, bats, CodeQL (actions+python), SonarCloud (Quality Gate passed, 0 new issues, 0 security hotspots), Secret scan (gitleaks), Agent Security Scan, unit-tests, Validate AW specs, AgentShield, validate-agent-profiles, gh-aw-compile. CodeRabbit was rate-limited (no findings); Copilot and Codex left no blocking concerns.
- INFO: Triage signal 'breaking configuration schema change' is inverted: the change moves aw.sh from reading non-existent flat keys (always empty) to reading the nested schema that actually exists in the workflow file. No existing workflow breaks; the 100% failure is fixed.
Reviewed by the PR-review cascade (triage: haiku 4.5 → deep: sonnet 4.6 + duck: o4-mini → audit: opus 4.7). Reply if you need a human review.



Closes #401
Implemented by dev-lead agent. Please review.