critique

Evaluate web interfaces for UX issues and generate prioritized P0–P3 remediation commands.

5|Updated Apr 18, 2026
One-click install
npx skills add https://github.com/russosalv/NONoise --skill critique-russosalv
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: critique
Source: https://github.com/russosalv/NONoise/tree/main/packages/skills/vendor/impeccable/critique
Command: npx skills add https://github.com/russosalv/NONoise --skill critique-russosalv

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Provide a structured, director-level evaluation of a user interface to uncover usability issues, AI-generated anti-patterns, cognitive overload, accessibility gaps, and priority fixes so teams can rapidly improve design quality and user outcomes.

Core Features & Use Cases

  • Dual assessment workflow: Runs an independent LLM-driven design review and a deterministic automated detector, then synthesizes results to reduce bias and surface both subjective and objective issues.
  • Heuristics & cognitive scoring: Produces Nielsen heuristic scores, an 8-item cognitive load summary, persona-driven red flags, and prioritized issues tagged P0–P3.
  • Actionable remediation: Maps findings to concrete fixes and recommended commands, highlights false positives, and offers a prioritized, repeatable plan for iterative improvement.
  • Use Case: Run this on a product landing page or component library to get a prioritized redesign plan, accessibility fixes, and a reproducible testing checklist for developers.

Quick Start

Ask the skill to review the target page or component and produce a combined LLM and automated critique with prioritized fixes.

Frequently Asked Questions about critique

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

FAQPage Schema
How do I run a UX audit on a web interface to find usability issues?

To run a UX audit, you can evaluate the web interface using a dual assessment workflow that combines an LLM-driven design review with a deterministic automated detector. This synthesizes subjective and objective findings to produce Nielsen heuristic scores, cognitive load counts, and prioritized P0–P3 issues.

What is the best way to check a design for accessibility gaps and cognitive overload?

Checking for accessibility gaps and cognitive overload involves assessing visual hierarchy and information architecture to generate an 8-item cognitive load summary and persona-driven red flags. This approach highlights specific barriers and maps findings to concrete fixes for iterative improvement.

Can I detect AI-generated design anti-patterns in my source markup?

Yes, you can detect AI-generated design anti-patterns by applying a combined review process to live pages or source markup. The evaluation identifies these specific anti-patterns across the interface and includes them in the prioritized remediation plan.

How do I prioritize UI fixes after a usability evaluation?

Prioritizing UI fixes after a usability evaluation involves tagging issues with P0–P3 severity levels. The evaluation maps these prioritized issues to concrete remediation commands and suggested fixes, offering a reproducible testing checklist for developers.

Does the UX critique process work on both live pages and design files?

Yes, the UX critique process works on both live web pages and design files. You can apply the assessment to pages, components, or source markup to evaluate information architecture, visual hierarchy, and accessibility regardless of the format.

What does a Nielsen heuristic scoring evaluation include for a component library?

A Nielsen heuristic scoring evaluation for a component library includes a structured assessment of usability heuristics, an 8-item cognitive load summary, and persona-driven red flags. It produces a reproducible testing checklist and a prioritized redesign plan.