# Review

> Check code for hallucinations and platform mismatches

- Skill: `rlaope/review` (Agent Skill)
- Install (CLI): `npx skillmds@latest add rlaope/review`
- Raw SKILL.md: https://api.skillmd.com/api/skills/rlaope/review/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: rlaope (https://skillmd.com/u/rlaope)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/rlaope/review

---


When this skill is invoked, IMMEDIATELY print:

```
[BW] scanning for hallucinations...
```

Run the review:
1. Check `git diff` (unstaged → staged → last commit) for:
   - Nonexistent imports (package not in dependencies)
   - Missing file references (relative imports to nowhere)
   - Platform-specific code on wrong OS
   - Wrong runtime APIs (Deno/Bun without the runtime)
   - Deprecated Node.js patterns
   - Suspicious AI-hallucinated method calls
   - Type safety bypasses (as any, ts-ignore)

2. Run `bash "${CLAUDE_PLUGIN_ROOT}/hooks/bestwork-review.sh"` or check manually.

After done, print:
```
[BW] review complete. {N} issues found.
```
or:
```
[BW] review complete. all clear — no hallucinations detected.
```