Menu
Review the last 24h of conversations and suggest improvements to CLAUDE.md, skills, and gotchas sections.
Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Basé sur la classification professionnelle SOC
When asked to review a PR or branch ("review this PR", "PR review", "argus", before merging), fan the diff out to parallel read-only reviewers and return a severity-ranked verdict; optionally post inline via gh.
When the user has reviewer feedback on a GitHub PR (human or bot like gemini-code-assist) and wants the pertinent items fixed — "j'ai des retours sur la PR", "résous/corrige ce qui est pertinent", "lis les commentaires de la PR via gh" — triage every comment thread via gh, apply only the pertinent fixes (drop nits), then reply to and resolve each triaged thread. Not for publishing our own review findings — that is argus-review --post.
Audit Claude Code agents, skills, and commands for quality and production readiness. Use when evaluating skill quality, checking production readiness scores, or comparing agents against best-practice templates.
Enforce project coding conventions: reduce duplication, keep naming/file placement consistent, and minimize regression risk. Includes React component and layout guidelines. Use when writing or reviewing code.
Generate a Product Requirements Document from user instructions using the PRD template. Outputs a human-readable self-contained HTML file plus a markdown source for implementation tooling. Includes a macro-level task checklist at the end.
After modifying code, map all consumers of changed symbols and flag unhandled mismatches. Not for debugging (use systematic-debugging).
| name | daily-review-prompt |
| description | Review the last 24h of conversations and suggest improvements to CLAUDE.md, skills, and gotchas sections. |
Analyze recent work sessions to identify friction points, missing context, and improvement opportunities for the developer experience.
Gather context from these sources (in order of priority):
git log --since="24 hours ago" --all --oneline across active projects~/.claude/projects/*/memory/~/.claude/CLAUDE.md) and per-project CLAUDE.md~/.claude/skills/ and ~/.claude/scheduled-tasks/ for what already existsFor each friction point, classify it:
| Type | Action |
|---|---|
| Missing context Claude needed | → CLAUDE.md update |
| Repeated workflow pattern | → New skill candidate |
| Existing skill that failed or was awkward | → Gotcha addition to that skill |
| Project-specific convention | → Project CLAUDE.md update |
Output a structured report — do not apply changes, only propose them.
## Summary
<!-- 2-3 sentence overview of what was worked on -->
## CLAUDE.md Changes
<!-- For each proposed change: what to add/modify and why -->
- **File**: `~/.claude/CLAUDE.md` or project-specific path
- **Change**: what to add or modify
- **Why**: what friction this would have prevented
## Skill Improvements
<!-- Gotchas to add to existing skills -->
- **Skill**: `<skill-name>`
- **Gotcha**: description of the failure/friction
- **Suggested addition**: text to add to ## Gotchas
## New Skills to Create
<!-- Only if a clear repeated pattern emerged -->
- **Name**: proposed skill name
- **Category**: one of Library/Verification/Data/Business/Scaffold/Quality/CI-CD/Runbook/Infra
- **Justification**: what repeated pattern this would address
## No Action Needed
<!-- Explicitly state if nothing worth changing was found — don't invent busywork -->
skill-creator if a new skill is approved for creationaudit-agents-skills to validate any skill changes meet quality barcoding-convention for code-related CLAUDE.md suggestions~/.claude/skills/ first to avoid duplicating existing ones