← ClaudeAtlas

review-commandslisted

Slash command prompt review - prompt quality, workflow completeness, cross-command consistency. Use for reviewing .claude/commands/*.md or .claude/skills/*/SKILL.md files.
mayankmankhand/llm-peer-review · ★ 7 · AI & Automation · score 74
Install: claude install-skill mayankmankhand/llm-peer-review
# Slash Command Review Be thorough but concise. **Use this when:** Reviewing slash command prompts (.claude/commands/*.md) - prompt quality, workflow completeness, cross-command consistency. **Don't use this when:** Reviewing application code (/review-code), testing a running web app (/review-browser), checking plan completion (/review-plan), evaluating end-user UX (/review-ux), or doing a pre-release check (/review-full). ## Critical Rules <rules> 1. **THE REVIEW PHASE REPORTS ONLY** - Reviewing never edits files; findings are its product. After the report, the same run continues into the auto loop (rule 2), which is what applies fixes 2. **Audit, then auto-fix, with pages** - Command-prompt findings are audited before the report per M2 in `.claude/skills/shared/hitl-loop.md`, so the report shows survivors only plus an Audited out log for the kills. They then enter the auto loop like any other (survivors auto-fixed, re-verified, then page/digest/log), with one built-in brake: fixing them means editing prompt files, which is an always-ask action (M9), so those fixes page for approval instead of auto-applying. Saying "report only" keeps a run report-first (M10) 3. **Explain simply** - Use plain English, avoid jargon </rules> ## How to Review <procedure> Read the command files being reviewed. Then pick one of two modes: **Small change** (1-2 files, minor wording tweaks): Review in a single pass. No sub-agents needed. **Bigger change** (3+ files or new/rewritten comma