autoclaw-reviewer

star 0

Self-review PANaCEa code changes. Check correctness, security, maintainability, performance, UX, edge cases, and test coverage. Use after every Builder sprint.

aaronjullger-lgtm By aaronjullger-lgtm schedule Updated 5/25/2026

name: autoclaw-reviewer description: Self-review PANaCEa code changes. Check correctness, security, maintainability, performance, UX, edge cases, and test coverage. Use after every Builder sprint. mode: reviewer

Reviewer Mode — Self-Critique

Purpose

Critique your own work before calling it done. Catch issues the Builder missed.

When to Use

  • After every Builder sprint
  • Before committing changes
  • Before claiming a task is complete
  • When reviewing sub-agent output

Review Checklist

Correctness

  • Does it solve the stated problem?
  • Are edge cases handled? (null, empty, error, boundary)
  • Is auth/ownership checked for data mutations?
  • Does data persist correctly?

Security

  • No secrets exposed in code or logs?
  • No process.env in Edge functions?
  • safePrismaDisconnect in finally blocks?
  • Input validated before DB writes?
  • User identity derived from auth, not client input?

Maintainability

  • Follows existing naming conventions?
  • No duplicated logic from existing code?
  • Imports clean (no unused)?
  • File structure matches repo patterns?

Performance

  • No N+1 queries?
  • No unnecessary re-renders in React?
  • No blocking operations without loading states?

UX

  • Loading state for async operations?
  • Empty state for no data?
  • Error state with user-actionable message?
  • Form submissions deduped?
  • No console errors?

Testing

  • New behavior covered by tests?
  • Edge cases tested?
  • Existing tests still pass?
  • No test ordering assumptions?

Coordination

  • Receives from: Builder (after every sprint), sub-agents (output verification)
  • Hands off to: Builder (if fixes needed), QA (if approved), Security (if risky changes detected)
  • Sub-agent review: Extra strict — verify build + tests, check import paths for wrong directory depth, check for missing ?. on injected deps

Sub-Agent Output Verification

npm run build          # Must pass
npm test               # 0 failures
rg "process\.env" functions/api/  # Edge hygiene check
rg "from '\.\.\/" --type ts | head -20  # Import depth sanity check

Common Pitfalls

  • Rubber-stamping: Review every dimension — don't skip security for "simple" changes
  • Missing Edge rules: Always check for process.env, safePrismaDisconnect in new Edge code
  • Sub-agent trust: Sub-agents produce wrong import paths — verify before accepting

Output Format

## Review: {change description}
**Verdict:** ✅ Approved / ⚠️ Needs fixes / ❌ Blocked

### Issues Found
1. {issue} — {file:line} — {fix needed}

### Risk Assessment
- **Blast radius:** {files/features affected}
- **Data risk:** {data mutation concerns}
- **Rollback ease:** {easy/medium/hard}
Install via CLI
npx skills add https://github.com/aaronjullger-lgtm/PANaCEa --skill autoclaw-reviewer
Repository Details
star Stars 0
call_split Forks 0
navigation Branch main
article Path SKILL.md
More from Creator
aaronjullger-lgtm
aaronjullger-lgtm Explore all skills →