audit

Run all quality gate agents on the codebase: code review, security, performance, silent failures, comments.

SKILL.md

GitHub

/audit — Full Quality Audit

Run every quality gate agent:

  1. Code review — agents/code-reviewer.md
  2. Language review — .ts/.tsx/.js/.jsx → agents/typescript-reviewer.md | .py → agents/python-reviewer.md
  3. Security — agents/security-reviewer.md (OWASP, secrets, deps)
  4. Silent failures — agents/silent-failure-hunter.md (empty catches, swallowed errors)
  5. Comments — agents/comment-analyzer.md (accuracy, staleness, TODOs)
  6. Performance — agents/performance-optimizer.md (bundle, algorithms, rendering)

Output: Per-category PASS/WARN/FAIL + overall verdict with severity counts.