review-ux

Audit application UX in a real browser and output structured findings with evidence.

Updated Jun 3, 2026
One-click install
npx skills add https://github.com/0xdeafcafe/skills --skill review-ux-0xdeafcafe
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: review-ux
Source: https://github.com/0xdeafcafe/skills/tree/main/skills/review-ux
Command: npx skills add https://github.com/0xdeafcafe/skills --skill review-ux-0xdeafcafe

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill helps you verify whether a feature actually feels correct in the browser before shipping, catching broken flows, missing loading states, accessibility issues, visual regressions, and confusing interaction patterns that code review alone often misses.

Core Features & Use Cases

  • Browser-driven UX review: Launches the app or uses a trusted preview URL, then walks the feature in a real browser using route and scenario-based testing.
  • Evidence-backed findings: Captures screenshots, checks console and network failures, and reports issues in a structured findings format suitable for PR review and follow-up.
  • Focused read-only auditing: Stays strictly non-destructive, making it ideal for validating the UX of a PR, smoke-testing a changed route, or reviewing a feature walkthrough without editing code.
  • Use Case: After opening a PR for a new checkout flow, use this Skill to click through the purchase path, test empty and error states, inspect Lighthouse regressions, and produce actionable findings with screenshots for the team.

Quick Start

Ask the agent to use review-ux to walk through the changed checkout flow in the current PR and report any UX findings with screenshot evidence.

Frequently Asked Questions about review-ux

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

FAQPage Schema
How do I validate UX and accessibility changes in a pull request before merging?

Perform a browser-driven UX review by walking the application's golden-path and edge-case states. This process captures screenshots, inspects console and network failures, and outputs actionable findings in a structured format suitable for pull request validation.

Can I run Lighthouse audits and capture screenshots during automated browser testing?

Yes, automated browser testing can run Lighthouse audits and capture screenshots. It exercises feature routes in a real browser to inspect accessibility, loading, and error states, producing structured findings with visual evidence for review.

What is a read-only browser audit for feature walkthroughs?

A read-only browser audit for feature walkthroughs is a non-destructive method that navigates application routes to verify user experience. It inspects interaction patterns, edge-case states, and accessibility regressions without modifying any code.

Does this automated UX review work for testing empty and error states?

Yes, automated UX review works for testing empty and error states. It applies route and scenario-based testing in a real browser to verify edge-case conditions, capturing structured findings and screenshot evidence for each state.

How do I verify console and network failures during a route walkthrough?

Verify console and network failures during a route walkthrough by using read-only browser automation. The audit inspects network requests and console errors while exercising the feature, documenting any failures in a structured findings output.