# Forge Review

> Use when the user asks to review work, a PR/MR, or whether a task is done in a repo containing `docs/forge/` or governed by FORGE. Runs the critique, security review, and evaluation gates in order; CI alone is not a review.

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

---


# FORGE Review

Use this skill when the user asks to review a FORGE task, PR, MR, branch, or
completed implementation.

## Workflow

1. Run `forge-critique` for scope, architecture, docs, contracts, and task
   metadata.
2. Run `forge-security-review` when the change surface or project policy
   requires security review.
3. Run `forge-evaluation` only after critique and required security review are
   complete.
4. Keep findings first and evidence-backed.

## Hard Stops

Stop when critique finds blockers, security review has unresolved concerns, or
evaluation evidence is incomplete.

## Evidence Required

- critique result
- security-review result or explicit n/a rationale
- evaluation gate result