Uiux frontend engineering
Use when designing, reviewing, implementing, or validating UI/UX and frontend experiences for web, mobile, responsive apps, design systems, Figma-to-code handoff, frontend stack/library selection, personas, UX mapping, information architecture, UI patterns, accessibility, interaction design, UX research, usability testing, theme/token application, or XR/spatial interfaces.From its SKILL.md
npx -y skills add oghie/skillsets --skill uiux-frontend-engineeringAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- 8 stars8 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
- runs commandsInstructs the agent to run 2 commands, including `scripts/uiux_static_audit.py <file-or-dir>` and 1 more.
What its file declares
Copied from the file, not written here
The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.
SKILL.md
8.3 KB, ~1.7k tokens by cl100k_base, as published. Nobody here has run it
UIUX Frontend Engineering
Core Rule
Treat UI as a tested product interface, not decoration: connect user goals, business goals, interaction model, visual system, implementation constraints, and validation evidence before shipping screens or components.
First Pass
- Classify the task: UX strategy, research, persona, UX mapping, IA/sitemap, user flow, wireframe, prototype, mockup, UI pattern selection, visual UI, design system, frontend stack/library selection, frontend implementation, responsive/mobile, accessibility, redesign/review, handoff, or XR/spatial interface.
- Identify target surface: web, mobile web, native mobile, desktop web app, kiosk, embedded screen, WebXR, VR, AR, MR, or spatial computing.
- Name users, context, primary jobs, failure modes, conversion or task metric, platform constraints, input method, and trust/safety concerns.
- Decide fidelity: sketch/flow, low-fidelity wireframe, clickable prototype, high-fidelity visual, coded prototype, production component, or XR graybox.
- Select validation before implementation: heuristic review, accessibility audit, responsive screenshots, usability test, click/first-impression test, A/B test, performance check, or XR comfort review.
Required Reads By Task
- End-to-end UX process, discovery, personas, user interviews, analytics, card sorting, surveys, support tickets, wireframes, validation, handoff, or QA:
references/ux-process-and-research.md. - Persona strategy, "Comparing Key Persona Types", buyer/customer/user/proto-persona distinction, goal-directed/role-based/fictional/engaging personas, customer journey maps, empathy maps, experience maps, service blueprints, touchpoints, or friction mapping:
references/personas-and-ux-mapping.md. - Information architecture, IA principles, taxonomy, labels, navigation models, search, card sorting, tree testing, HTML/XML/mobile/image/video/news sitemaps, or sitemap handoff:
references/information-architecture-and-sitemaps.md. - Wireframes, prototypes, mockups, fidelity choice, paper/Wizard-of-Oz/clickable/coded prototypes, microcopy, microinteractions, or mockup-to-implementation readiness:
references/wireframes-prototypes-and-mockups.md. - UI pattern selection, visual hierarchy patterns, Z-pattern/F-pattern layouts, primary actions, forgiving input, breadcrumbs, progressive disclosure, lazy registration, infinite scroll, modals, cards, autocomplete, sticky navigation, mobile UI patterns, or dark-pattern review:
references/ui-patterns-catalog.md. - Psychology laws, cognitive load, target sizing, choice overload, familiarity, aesthetics, memory, complexity, responsiveness, ethics, or dark-pattern risk:
references/psychology-and-interaction-laws.md. - Visual UI composition, Gestalt, hierarchy, color, typography, spacing, affordance, signifiers, forms, cards, states, icons, illustrations, or motion:
references/visual-interface-composition.md. - Figma, design systems, tokens, components, auto layout, constraints, responsive/adaptive design, plugin/handoff flow, or developer collaboration:
references/responsive-design-systems-and-handoff.md. - Frontend framework/library selection, styling approach, component library choice, Radix UI, shadcn/ui, Ant Design, Mantine UI, MUI, vanilla CSS, CSS Modules, Linaria, PostCSS, Bulma, Tailwind CSS, or stack trade-offs:
references/frontend-stack-and-library-selection.md. - Frontend implementation, accessibility, performance, state coverage, visual QA, component contracts, framework-neutral CSS/component mapping, or production readiness:
references/frontend-implementation-quality.md. - Mobile/touch design, one-handed use, touch targets, mobile navigation, forms, gestures, device preview, or native-app constraints:
references/mobile-and-touch-ux.md. - XR, spatial computing, WebXR, VR, AR, MR, 3D UI, spatial objects, OOUX, locomotion, comfort, haptics, voice, gaze, avatars, embodiment, safety, or spatial analytics:
references/xr-spatial-ux-and-3d-interfaces.md. - AI-assisted design, generative assets, design copilots, AI disclosure, sourcing, IP, safety, privacy, transparency, or manipulation risk:
references/ai-assisted-and-ethical-design.md. - Theme/style selection, static token database, CSS variables, framework/library token mapping, or visual style guardrails:
references/theme-token-database.mdand the relevant file undertheme/.
Task Playbooks
- Build or redesign a product UI:
tasks/product-ui-build.md. - Convert theme/style into implementation tokens and components:
tasks/design-system-token-implementation.md. - Review an existing UI/UX or frontend:
tasks/usability-review-and-redesign.md. - Build responsive or mobile-first frontend:
tasks/responsive-mobile-frontend.md. - Design or review XR/spatial interface:
tasks/xr-spatial-interface-design.md. - Prepare prototype validation, handoff, and QA:
tasks/prototype-validation-handoff.md.
Theme Contract
- Do not modify
theme/unless the user explicitly asks for theme authoring. - Treat
theme/index.yamlas the registry and each style file as the source of truth. - Read
theme/_schema/style.schema.yamlonly when field meaning is unclear. - Prefer a style because its
use_cases,mood,keywords, andavoid_whenmatch the product, not because it looks novel. - Map tokens directly into implementation variables; do not invent a parallel palette, radius scale, spacing scale, or motion personality.
Design Discipline
- Start with the user's task and failure path, then choose UI structure.
- Prefer the existing project stack when it is coherent; add or change UI libraries only after naming the accessibility, theming, delivery-speed, density, bundle, SSR, and maintenance trade-offs.
- Use familiar patterns by default; depart from convention only when the new pattern improves the core task and can be tested.
- Design every interactive component with states: default, hover, focus, active, disabled, loading, empty, error, success, and reduced-motion where relevant.
- Make accessibility structural: semantic markup, keyboard path, target size, contrast, labels, focus order, captions/alternatives, motion controls, and screen-reader names.
- Treat responsive design as layout behavior, not breakpoint cosmetics: define constraints, grid, content priority, reflow rules, and overflow behavior.
- For XR, design objects, spatial relationships, interaction distance, comfort, onboarding, and safety before polishing panels.
- For AI-generated UI or assets, document sourcing, disclosure, rights risk, user trust, and human review.
Visual And Diagram Standards
- Use diagrams to clarify process, hierarchy, interaction, object relationships, state transitions, or validation loops.
- Prefer reusable templates in
templates/for UX process, validation loop, persona comparison, UX mapping, IA/sitemap flow, UI pattern decisions, responsive token mapping, component state matrix, and spatial object maps. - For frontend work, verify rendered output with screenshots at mobile and desktop widths; for XR/WebGL/WebXR, verify the scene is nonblank, framed, and interactive.
Script Helpers
- Run
scripts/uiux_static_audit.py <file-or-dir>to scan design plans, UI specs, PR notes, or Markdown docs for missing users, flows, accessibility, responsive behavior, states, validation, handoff, ethics, and XR concerns. - Run
scripts/theme_catalog.py --root uiux_frontend_skill/theme --listto list available style ids, or filter with--categoryand--keywordbefore selecting a theme.
Output Standard
Lead with the design judgment or implementation path. State target surface, users, core task, chosen process depth, style/theme decision, UX laws or heuristics used, component/state coverage, responsive/mobile/XR handling, validation checks, implementation risks, and what remains uncertain.
What ships with it: 61 files
336.1 KB alongside SKILL.md, 2 of them executable
agents/
- openai.yaml353 B
references/
- ai-assisted-and-ethical-design.md2.5 KB
- frontend-implementation-quality.md3.4 KB
- frontend-stack-and-library-selection.md6.4 KB
- information-architecture-and-sitemaps.md4.0 KB
- mobile-and-touch-ux.md2.1 KB
- personas-and-ux-mapping.md5.5 KB
- psychology-and-interaction-laws.md4.1 KB
- responsive-design-systems-and-handoff.md4.3 KB
- theme-token-database.md3.8 KB
- ui-patterns-catalog.md5.1 KB
- ux-process-and-research.md7.5 KB
- visual-interface-composition.md4.1 KB
- wireframes-prototypes-and-mockups.md3.9 KB
- xr-spatial-ux-and-3d-interfaces.md4.1 KB
scripts/
- theme_catalog.pyruns4.3 KB
- uiux_static_audit.pyruns6.2 KB
tasks/
templates/
theme/
- agent.md4.5 KB
21 more files not listed here. See all 61 in the repository.
Gives 0 of the 12 instructions most ux research skills give in ~1.7k tokens
Counted across 242 of the 261 authors here whose files we hold, read 2026-09-06
- Fall back to prompt text when image generation failsin 16 of 242, across 7 files
- Offer three name candidates with reasonsin 16 of 242, across 7 files
- Comment on each result before asking the userin 16 of 242, across 7 files
- Check for an approved image-generation skill before auto-generatingin 15 of 242, across 6 files
- Degrade gracefully instead of interrupting on errorsin 14 of 242, across 5 files
- Guide the user to write SOUL.md and IDENTITY.mdin 13 of 242, across 6 files
- Derive two to four boundary rules in character voicein 11 of 242, across 4 files
- Report SXO score separately from SEO Health Scorein 11 of 242, across 6 files
- Run gacha.py for random drawsin 11 of 242, across 4 files
- Score the page across seven gap dimensionsin 11 of 242, across 6 files
- Derive three to five user stories citing SERP signalsin 11 of 242, across 6 files
- Confirm DataForSEO cost estimates before calling its APIsin 11 of 242, across 6 files
Said here and by no other author read
- Classify the task and target surface first
- Name users, context, jobs, and failure modes
- Choose validation method before implementing
- Start with the user's task and failure path
- Prefer the existing coherent project stack
- Make accessibility structural with semantic markup
Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.