pr-review-ci-loop-9478a3 — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited pr-review-ci-loop-9478a3 (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Combine PR review and CI failure triage into one human-controlled loop. Review findings and CI logs are inputs; they are not permission for autonomous remote actions.
Do not use when the user only asked for a review, only asked for CI diagnosis, or has not approved edits.
gh for GitHub-hosted PRsgh pr view <pr> --json number,title,author,baseRefName,headRefName,mergeStateStatus,reviewDecision,statusCheckRollup
gh pr diff <pr> --stat
gh pr diff <pr>Record file count, risky areas, generated files, and public API changes.
Classify findings by severity:
| Category | What to check |
|---|---|
| Correctness | Logic errors, off-by-one, wrong assumptions |
| Tests | Missing coverage for changed behavior, broken assertions |
| Public API | Breaking changes to exports, signatures, or contracts |
| Security | Injection risks, exposed secrets, auth bypass |
| Migration / release risk | Schema changes, feature flags, rollback difficulty |
gh pr checks <pr>
gh run view <run-id> --log-failedExtract the first failing assertion or error line. Categorize each failure as code error, flaky, environment, or config.
REVIEW FINDINGS:
[blocking] <description> - <file:line or section>
[non-blocking] <description>
CI FAILURES:
<check name>: <first error line>
Likely cause: <code error / flaky / env / config>
LIKELY FIXES:
1. <specific change>
2. <specific change>
VALIDATION PLAN:
- Run: <command>
- Expected: <outcome>State the exact operation before running it:
NEXT REMOTE ACTION: git push origin <branch>
EFFECT: updates the remote branch, triggering CI re-run
Approve?~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.