# Critic

> Invokes the critic-agent to review the latest produced output. Use when the user runs /critic to request a quality review of the most recent artefact or deliverable from the current workflow.

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

---


# Critic

1. **Identify the latest output** — the most recent artefact produced in the current workflow
2. **Invoke the critic-agent** with complete context: the output file(s), the brief or requirements the output was produced against, applicable standards, and relevant background — what the artefact is for, where it fits, what problem it solves
3. **Report the verdict** — share the critic's accept/revise decision and key findings with the user

The critic-agent works in isolation — it only knows what you give it. Provide enough for an informed review: the brief, the standards, and the purpose, not just the file.