ux-audit

Audits live web-platform UI by capturing screenshots and extracting structured findings.

12|2|Updated Jan 27, 2026
One-click install
npx skills add https://github.com/jikig-ai/soleur --skill ux-audit-jikig-ai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ux-audit
Source: https://github.com/jikig-ai/soleur/tree/main/plugins/soleur/skills/ux-audit
Command: npx skills add https://github.com/jikig-ai/soleur --skill ux-audit-jikig-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

The ux-audit skill automates recurring visual and usability reviews of a live web platform to detect UI decay, surface actionable findings, and prevent regressions from accumulating unnoticed. It reduces manual screenshot review, prevents duplicate bug filings, and ensures the highest-severity UX problems on primary product surfaces are prioritized for the team.

Core Features & Use Cases

  • Scheduled and on-demand audits across authenticated dashboard routes and anonymous funnel pages with configurable per-route viewport sizes.
  • Automated screenshot capture via Playwright with bot-authenticated sign-in and fixture seeding to ensure reproducible states.
  • Delegated design analysis: hands off screenshots to the ux-design-lead agent in audit mode and parses JSON findings.
  • Deterministic deduplication and capping using SHA-256 finding hashes, global/per-run/per-route caps, and GitHub issue search to avoid duplicates.
  • Dry-run and file modes: dry-run writes findings JSON/artifacts while file mode creates and attaches markdown issue bodies and screenshots via the GitHub CLI.
  • Use Case: Run monthly audits to capture regression drift on the founder dashboard, automatically surface the top-priority UX issues, and file a small, curated set of issues for the design/product team.

Quick Start

Run a scheduled or local dry-run ux-audit to capture screenshots, analyze them via the ux-design-lead agent, and emit a findings JSON for review.

Frequently Asked Questions about ux-audit

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

FAQPage Schema
How do I automate UX audits to detect visual decay on my web platform?

Automate recurring UX audits by capturing screenshots via Playwright across authenticated dashboard routes and anonymous funnel pages, delegating analysis to a design agent, and enforcing per-route caps to detect visual decay. It runs scheduled or ad-hoc reviews to surface regressions.

What is UI decay detection and how does automated screenshotting work?

UI decay detection identifies visual regressions and usability issues on live web platforms. Automated screenshotting works by using Playwright with bot-authenticated sign-in and fixture seeding to capture reproducible states across configurable viewports for design analysis.

How do I prevent duplicate GitHub issues when filing automated UX findings?

Prevent duplicate GitHub issues by using deterministic deduplication with SHA-256 finding hashes and GitHub CLI issue search before filing. The audit enforces per-run, per-route, and global caps to ensure only a curated set of high-severity issues are created.

Can I run a UX audit in dry-run mode without creating GitHub issues?

Yes, you can run a UX audit in dry-run mode. Dry-run mode captures screenshots, analyzes them via the design agent, and writes findings JSON and artifacts locally for review without using the GitHub CLI to create or attach issues.

Does the automated UX audit support authenticated dashboard routes?

Yes, the automated UX audit supports authenticated dashboard routes. It uses Supabase bot-authenticated sign-in and Doppler-provided environment variables to access secure areas alongside anonymous funnel pages during scheduled or ad-hoc screenshot capture.

What are the limitations of capping UX audit findings per route?

Capping UX audit findings per route limits the number of issues filed to prevent overwhelming the team. It uses global, per-run, and per-route caps to ensure only the highest-severity UX problems on primary product surfaces are prioritized.