critique

Identify and report structural flaws in plans, designs, or code through adversarial review.

14|1|Updated Mar 28, 2026
One-click install
npx skills add https://github.com/dlowd/claude-skill-critique --skill critique-dlowd
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: critique
Source: https://github.com/dlowd/claude-skill-critique/tree/main
Command: npx skills add https://github.com/dlowd/claude-skill-critique --skill critique-dlowd

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill provides a skeptical, adversarial review of plans, design docs, code, or projects to identify missing considerations, weak assumptions, and risky design choices.

Core Features & Use Cases

  • Adversarial critique of architecture and implementation to surface showstoppers, gaps, inconsistencies, and underspecified areas.
  • Reads project context and target artifacts to generate actionable risk-focused feedback for solo developers or small teams.
  • Produces prioritized findings with concrete scenarios to illustrate impact.

Quick Start

Provide CLAUDE.md context and a target artifact to generate a risk-focused adversarial review.

Frequently Asked Questions about critique

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I perform an adversarial security review of my architecture design?

An adversarial security review identifies fundamental flaws in architecture by evaluating project context to surface showstoppers, gaps, and inconsistencies. It generates structured findings with concrete scenarios to highlight risky design choices and missing considerations.

What is an adversarial code review and when do I need it?

Adversarial code review is a skeptical evaluation of code to find weak assumptions and risky choices. You need it to catch showstoppers and underspecified areas before implementation, ensuring your design documents and proposals handle edge cases properly.

How do I find fundamental flaws and gaps in my project plans?

To find fundamental flaws and gaps in project plans, perform a skeptical review against repository context. This process categorizes findings into showstoppers, inconsistencies, and underspecified areas, providing actionable scenarios with precise root causes for each risk.

Can I use adversarial review for solo developer projects?

Yes, adversarial review works for solo developer projects by reading project context and target artifacts to generate risk-focused feedback. It acts as a skeptical reviewer to surface missing considerations and weak assumptions that a solo developer might overlook.

What is the best way to review code for security risks and design inconsistencies?

The best way to review code for security risks and design inconsistencies is an adversarial review that prioritizes showstoppers and gaps. This approach produces structured findings with concrete scenarios, illustrating the impact of weak assumptions and risky design choices.

What are the limitations of using an adversarial review for codebase analysis?

Adversarial review for codebase analysis relies on existing project context and target artifacts to identify flaws. If the repository context is incomplete or the design documents are missing critical details, the review may not surface all relevant underspecified areas and inconsistencies.