Critique — Development skill for Claude Code
Verify that a plan was fully implemented — checks acceptance criteria against code.
How to install Critique
Installs to ~/.claude/commands/cognitx-leyton-graphrag-code-critique.md
mkdir -p ~/.claude/commands && curl -fsSL https://raw.githubusercontent.com/cognitx-leyton/graphrag-code/HEAD/.claude/commands/critique.md -o ~/.claude/commands/cognitx-leyton-graphrag-code-critique.md Restart Claude Code, or start a new session, for it to be picked up.
What Critique does
description: Verify that a plan was fully implemented — checks acceptance criteria against code argument-hint: allowed-tools: Read, Grep, Glob, Bash(.venv/bin/*), Bash(python:*), Agent
Critique (Step 7)
**Plan**: $ARGUMENTS
Verify the plan was fully and correctly implemented. This is NOT a code review (that's `/review-pr`) — this checks **completeness** against the plan's specification.
Phase 1: Load Plan
Read the plan and extract:
- Tasks — every task lis
Alternatives in Development
- Implement Plan — You are tasked with implementing an approved technical plan from thoughts/shared/plans/ 10k ★
- Product Manager Skills — PM skill for Claude Code, Codex, Cursor, and Windsurf: diagnose SaaS metrics, critique PRDs, plan roadmaps, ru 149 ★
- Verify — Verify frontend changes against acceptance criteria using browser agents 107 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Ac Validation
Review Acceptance Criteria for one or more user stories against completeness and quality standards. Checks hap
Work Release
Coordinate a release session — confirm scope, brief and run the required agents, verify against acceptance cri
Agent Verify
SDD Phase 4 — Run acceptance-criteria verification for all [x] tasks in current batch. Produce VERIFICATION_RE
Sdlc Code Implementation
Build working software that meets acceptance criteria. Use when executing implementation plans. Triggers: impl
Orbits Brief
Write a self-contained implementer brief for one chunk of an approved ORBITS plan — with the source material p
Adw Plan
ADW step 2 - turn a scouted run into a plan with testable acceptance criteria
Related Agents
Gantry Critic
Attacks a plan before it is implemented. Use to grill plan.md against task.md — unstated assumptions, unfalsif
Harness PM
Product manager — researches the codebase and plans in one context, writing BRIEF.md and plan.yaml with fully
Verify Agent
Verifies a plan has no loose ends before it gets implemented. Checks every step against the real codebase for