mcpField
U
ux-audit-skill

ux-audit-skill

github.com/sergekostenchukDesign & UI3263 installsUpdated 18w ago
Repository

About ux-audit-skill

ux-audit-skill reviews existing interfaces using available evidence and produces a severity matrix with actionable refactoring guidance. It supports audits of live URLs, screenshots, and source code while distinguishing evidence types and limits.

Key features

  1. 01Review live URLs, screenshots, and code
  2. 02Produce severity matrices and fix guidance
  3. 03Label the evidence behind each finding
  4. 04Compare before-and-after evidence for regressions

Use cases

  1. 01Assess interface issues from screenshots
  2. 02Review an existing page or source files
  3. 03Turn audit findings into implementation tasks
  4. 04Check regression risks after changes

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/sergekostenchuk/ui-ux-agent-skill-system --skill ux-audit-skill

After installing, type/ux-audit-skillin Claude Code to confirm it loaded.

Bundled files

References and scripts installed along with SKILL.md.

agents/
openai.yaml
assets/
audit-report.template.md
regression-report.template.md
evals/
evals.json
references/
audit-evidence-model.md
mobile-touch-rubric.md
severity-model.md
wcag-quick-ref.md

Details

From the project README

UX Audit Skill

Use this specialist for evidence-backed review of existing UI artifacts. The output is a severity matrix and actionable refactor brief, not a generic design essay.

Operating Modes

  • review-live-url: audit a reachable URL with available browser/static evidence.
  • review-screenshot: audit image-only context without DOM claims.
  • review-code: audit source files and templates.
  • refactor-brief: convert findings into implementation tasks for another skill.
  • regression-check: compare before/after evidence and remaining risks.

Required Workflow

  1. Classify available evidence using references/audit-evidence-model.md.
  2. Apply references/severity-model.md before ranking findings.
  3. Use assets/audit-report.template.md for audit output.
  4. Use assets/regression-report.template.md for before/after checks.
  5. Apply ../../shared/privacy-policy.md before screenshots, Figma, or authenticated state.
  6. If a URL is available, reuse local scripts listed in ../../shared/tool_inventory.json.
  7. Label every finding by evidence type: DOM, screenshot, source, automated, or Manual.

Evidence Boundaries

  • Screenshot-only input cannot support DOM, Lighthouse, axe, or runtime claims.
  • Source-only input cannot prove rendered layout unless run or screenshot evidence exists.
  • Manual findings are allowed but must be labeled Manual.
  • Planned tools are not evidence.

Validation

python3 $CODEX_SKILLS_DIR/.system/skill-creator/scripts/quick_validate.py $CODEX_SKILLS_DIR/ux-audit-skill
python3 $CODEX_SKILLS_DIR/senior-skill-architect/scripts/lint_production_skill.py $CODEX_SKILLS_DIR/ux-audit-skill

Output Shape

Audit target:
Evidence available:
Ran:
Skipped:
Severity matrix:
Top fixes:
Verification plan:
Privacy notes:
Residual risk:

Comments

Markdown supported