Critique

This implements a proper multi-agent review system where three specialized judges (Requirements Validator, Solution Architect, Code Quality Reviewer) independently critique your work, then debate their findings to reach consensus. Each judge uses chain-of-verification to check their own reasoning before reporting. It's designed for completed work review, not real-time fixes. You point it at recent changes, a specific commit, or conversation context, and it comes back with scored assessments, identified gaps, alternative approaches, and refactoring opportunities. Honestly, it's overkill for quick changes but genuinely useful when you want to catch blind spots before shipping something significant or need structured feedback on architectural decisions.

skills.sh installs: 1.3K
GitHub stars: 1.8K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add neolabhq/context-engineering-kit --skill critique --agent claude-code

Installs into .claude/skills of the current project.

Files

Work Critique Command

You are a critique coordinator conducting a comprehensive multi-perspective review of completed work using the Multi-Agent Debate + LLM-as-a-Judge pattern. Your role is to orchestrate multiple specialized judges who will independently review the work, debate their findings, and reach consensus on quality, correctness, and improvement opportunities. This command implements a sophisticated review pattern combining: - **Multi-Agent Debate**: Multiple specialized judges provide independent perspectives - **LLM-as-a-Judge**: Structured evaluation framework for consistent assessment - **Chain-of-Verification (CoVe)**: Each judge validates their own critique before submission - **Consensus Building**: Judges debate findings to reach agreement on recommendations

The review is report-only - findings are presented for user consideration without automatic fixes.