critique

Run dual LLM and automated UX critiques against Nielsen heuristics.

184|16|Updated Mar 22, 2025
One-click install
npx skills add https://github.com/zhushen12580/smart-screenshot --skill critique-zhushen12580
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: critique
Source: https://github.com/zhushen12580/smart-screenshot/tree/main/.claude/skills/critique
Command: npx skills add https://github.com/zhushen12580/smart-screenshot --skill critique-zhushen12580

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides a thorough UX critique by coordinating an LLM-based design review and an automated detection pass to produce actionable feedback, scores, and prioritized issues for design critique tasks.

Core Features & Use Cases

  • Dual assessments: a detailed LLM-driven evaluation of visual hierarchy, information architecture, and overall quality, plus an automated detector scan for AI-generated design patterns.
  • Combined reporting: a synthesized critique with a Design Health Score, Nielsen heuristic scores, priority issues, and persona-based red flags.
  • Structured guidance: outputs include actionable fixes and recommended next steps for design teams and product managers.

Quick Start

Provide the design assets (URL or screenshots) and run the critique workflow to receive a combined UX critique.

Frequently Asked Questions about critique

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

FAQPage Schema
How do I perform a structured UX critique on a live web page?

A structured UX critique evaluates visual hierarchy and information architecture using dual assessments. It combines an LLM-driven design review with automated detection to produce a Design Health Score, prioritized issues, and persona red flags from your provided assets.

Can I evaluate my UI against Nielsen heuristics and cognitive load guidelines?

Yes, you can evaluate your UI against Nielsen heuristics and cognitive load guidelines. The skill runs a detailed LLM-driven evaluation to score your design and identify actionable fixes based on these specific usability principles.

What is the best way to identify priority UX issues and persona red flags in a design?

The best way to identify priority UX issues and persona red flags is through a dual assessment process. An LLM design review combined with automated detection unifies findings into a synthesized critique highlighting critical flaws for specific user personas.

Does the UX critique workflow support browser automation for evaluating live pages?

Yes, the UX critique workflow supports browser automation for evaluating live pages. This allows the automated detection pass to scan live URLs directly, identifying AI-generated design patterns and usability flaws without requiring manual screenshots.

How do I get actionable design feedback from UI screenshots?

To get actionable design feedback from UI screenshots, you provide the image assets to the critique workflow. The system analyzes them to return structured guidance including actionable fixes, recommended next steps, and suggested commands for refinement.