From ux-pro
This skill should be used when the user asks to "audit accessibility", "evaluate usability", "review UX", "check WCAG compliance", "design a user flow", "plan user research", "create a test script", "write a journey map", "build a service blueprint", "improve navigation", "fix error messages", "write UI copy", "add ARIA", "check keyboard accessibility", "design a form", "plan information architecture", "conduct a heuristic evaluation", "measure UX metrics", "write microcopy", or mentions WCAG, ARIA, Nielsen heuristics, Laws of UX, usability testing, card sorting, tree testing, SUS score, HEART framework, service design, content design, interaction patterns, focus management, screen reader, or any UX research method. Covers the full UX discipline: research, information architecture, interaction design, accessibility, usability testing, content design, and service design.
How this skill is triggered — by the user, by Claude, or both
Slash command
/ux-pro:ux-expertiseThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Foundational UX knowledge covering research methods, information architecture,
references/aria-patterns.mdreferences/inclusive-design.mdreferences/information-architecture.mdreferences/interaction-patterns.mdreferences/laws-of-ux.mdreferences/nielsen-heuristics.mdreferences/research-methods.mdreferences/service-design.mdreferences/ux-metrics.mdreferences/ux-writing-guide.mdreferences/wcag-22-design-guide.mdFoundational UX knowledge covering research methods, information architecture, interaction design patterns, accessibility standards, usability testing, content design, and service design. This skill provides the knowledge layer that all seven specialist agents draw from.
The primary evaluation framework for heuristic reviews:
Load references/nielsen-heuristics.md for per-heuristic checklists and examples.
Psychological principles for design decisions. Key laws:
Load references/laws-of-ux.md for all 17 laws with origins and applications.
Accessibility requirements organized by design responsibility:
| Category | Key criteria |
|---|---|
| Visual design | Contrast 4.5:1 text, 3:1 UI; no color-only info; target 24x24px min |
| Interaction | Keyboard access all; no traps; focus visible; consistent nav |
| Content | Descriptive titles; heading hierarchy; labeled inputs; clear errors |
| Structure | Landmarks; reading order = DOM order; works portrait and landscape |
Load references/wcag-22-design-guide.md for the complete organized checklist.
Choose methods by project phase:
| Phase | Methods | Effort |
|---|---|---|
| Discovery | Contextual inquiry, stakeholder interviews, diary studies, competitive analysis | Medium-High |
| Definition | Personas, journey mapping, JTBD, service blueprints, task analysis | Medium |
| Design | Usability testing, A/B testing, tree testing, heuristic evaluation, cognitive walkthrough | Low-Medium |
| Measurement | Analytics review, SUS, HEART framework, task-level metrics | Low |
Load references/research-methods.md for the complete framework with method details.
The four systems of IA (Rosenfeld and Morville):
Navigation models: Hub and Spoke, Nested Doll, Tabbed View, Bento Box, Filtered View.
Load references/information-architecture.md for frameworks, card sorting, tree testing.
Key pattern categories: Form Design, Navigation, Data Display, Feedback and Messaging, Onboarding, Complex Interactions (drag/drop, inline edit, bulk actions).
Load references/interaction-patterns.md for the complete pattern catalog.
Keyboard interaction models for every common widget: Accordion, Combobox, Dialog, Listbox, Menu, Tabs, Tree View, and more.
Load references/aria-patterns.md for the complete pattern set with keyboard models.
| Framework | Dimensions |
|---|---|
| Task-level | Completion rate, time on task, error rate, lostness, efficiency |
| SUS | 10-item questionnaire, score 0-100, benchmark ~68 |
| HEART | Happiness, Engagement, Adoption, Retention, Task success |
| AARRR | Acquisition, Activation, Retention, Revenue, Referral |
Load references/ux-metrics.md for measurement frameworks and calculation methods.
Always verify accessibility patterns against real implementations:
# Radix: focus management and ARIA
grep -r "aria-\|role=\|tabIndex\|onKeyDown" refs/radix-primitives/packages/react/dialog/src/
# axe-core: WCAG rules as code
grep -r "description\|help\|tags" refs/axe-core/lib/rules/ --include="*.json"
# ARIA practices: keyboard models
grep -r "keyboard\|key.*down\|arrow" refs/aria-practices/content/patterns/
# GOV.UK: research-backed patterns
grep -r "research\|evidence\|finding" refs/govuk-design-system/src/patterns/
Reusable deliverable scaffolds in templates/:
templates/heuristic-evaluation/: Report with per-heuristic sections, severity scaletemplates/usability-test-script/: Moderator guide, tasks, metrics, SUS, debrieftemplates/research-plan/: Brief, questions, methods, participants, timelinetemplates/journey-map/: Phases, rows (actions/thoughts/feelings/touchpoints)templates/service-blueprint/: Layers (frontstage/backstage/support), lines of visibilitytemplates/accessibility-audit/: WCAG checklist by level, issue template, scoringtemplates/ux-review/: Screen-by-screen format, priority matrixreferences/nielsen-heuristics.md: 10 heuristics with checklists and violation examplesreferences/laws-of-ux.md: 17 UX laws with psychology origins and applicationsreferences/research-methods.md: Method selection by phase, study design guidancereferences/information-architecture.md: IA systems, navigation models, card sortingreferences/wcag-22-design-guide.md: WCAG 2.2 organized by designer responsibilityreferences/aria-patterns.md: Keyboard interaction models per widget typereferences/interaction-patterns.md: Form, navigation, data display, feedback patternsreferences/ux-metrics.md: HEART, SUS, task metrics, AARRR frameworksreferences/service-design.md: Journey maps, blueprints, empathy maps, JTBDreferences/ux-writing-guide.md: Microcopy patterns, voice/tone, content hierarchyreferences/inclusive-design.md: Microsoft framework, disability spectrum, universal designGuides creation, editing, and verification of skills for AI coding agents using test-driven development with subagent scenarios. Use when authoring or debugging skills.
npx claudepluginhub travis-gilbert/claude-marketplace --plugin ux-pro