ui-craft
Provides guardrails for user-facing UI work: usability heuristics, accessibility floors, design-system discipline, component states, microcopy, motion, dark mode, responsive behavior, and human-AI UX. Use when designing, generating, reviewing, or refactoring visible product surfaces such as components, pages, dashboards, forms, dialogs, loading/empty/error states, or AI interfaces. Do not use for backend-only work, infrastructure, CLI/TUI design, or pure documentation editing.
What this skill does
# UI Craft ## The Fundamental Law ``` AI-GENERATED UI FAILS THE SAME WAYS EVERY TIME. Visual sameness. Weak hierarchy. Fake interactivity. Half the states missing. Hex codes outside the token system. "Welcome!" microcopy that nobody asked for. This skill exists so the UI being shipped is not slop. ``` UI craft is the discipline of producing user-facing surfaces that respect the user, the brand, and the medium. It is not "make it pretty." It is: every visible state explicit, every interactive element reachable by keyboard, every value drawn from the design system, every word earning its place, every motion serving a purpose. A user interface is a contract with a real person. Ship slop and the contract breaks. ## Required Reading Router Match your task to the row. Read the listed files **in full before generating anything visual**. They are not appendices — they are load-bearing. Inline content in this SKILL.md is a pointer to the reference file, not a substitute for it. Files are loaded JIT (never all at once); but every row whose trigger applies to your task is mandatory, not optional. | Task | MUST read | | --------------------------------------------- | ---------------------------------------------------------------- | | Any interactive widget (dialog, menu, combo) | `references/accessibility-floor.md` | | Auditing AI-generated UI | `references/ai-slop-patterns.md` + `references/anti-defaults.md` | | Designing a novel surface | `references/usability-foundations.md` | | Picking or extending a component pattern | `references/component-patterns.md` | | Writing or reviewing user-visible text | `references/microcopy-quality.md` | | Implementing or reviewing dark mode | `references/dark-mode.md` | | Animation / transitions / motion review | `references/motion-patterns.md` | | AI / agent UI (chat, streaming, citations) | `references/human-ai-ux.md` | | Working with DESIGN.md or extending tokens | `references/design-system-integration.md` | | Performance work / pre-ship gates | `references/performance.md` | | "We need a [Linear/Notion-like] X" brief | `references/archetypes.md` | | No existing token system in the repo | `references/visual-craft.md` | ## Reference Index What each file covers (the Router above tells you *when* to load them): - `references/usability-foundations.md` — Nielsen 10 heuristics, Laws of UX, progressive disclosure, mental models. - `references/accessibility-floor.md` — Full WCAG 2.2 AA checklist, ARIA patterns (dialog, combobox, menu, tabs, slider, listbox), verification recipes. - `references/visual-craft.md` — Typography scales, color systems, spacing rhythm, motion physics, dark mode, elevation. - `references/component-patterns.md` — Do/don't per component: buttons, inputs, dialogs, navigation, tables, dropdowns, tabs, popovers, toasts. - `references/microcopy-quality.md` — Full banned-vocabulary list, tone guides per surface, error/empty/CTA copy rules. - `references/ai-slop-patterns.md` — Before/after for each of the 14 anti-slop patterns. Detection prompts. Remediation recipes. - `references/anti-defaults.md` — 17 literal artifacts to refuse on sight (emoji-as-icon, "Inter" by reflex, centred hero, placeholder names, gradient text, glassmorphism padrão, modal-as-first-thought, etc.). - `references/human-ai-ux.md` — Microsoft + IBM agent/AI UI guidelines distilled. - `references/design-system-integration.md` — Canonical DESIGN.md sections, semantic tokens, shadcn/Radix composition, rules-files for AI codegen. - `references/performance.md` — 80ms perceived-performance threshold, font-loading zero-CLS strategy, executable pre-ship gate. - `references/motion-patterns.md` — 100/300/500 duration rule, easing curves, Tailwind/CSS/Framer code, reduced-motion templates, severity-tagged anti-patterns. - `references/dark-mode.md` — Surface-lightness elevation, accent desaturation, semantic token strategy, dangerous combinations. - `references/archetypes.md` — Seven named archetypes (Bento, Command Palette, Focus Mode, Live Status, Intelligent List, Empty State Hero, Inline Editor) each with a six-slot signature contract. ### Scripts - `scripts/check-contrast.mjs` — verify foreground/background WCAG AA/AAA + APCA Lc. Zero-dependency, accepts hex / rgb / oklch. Wire into CI when token files change. - `scripts/detect-token-drift.mjs` — scan source for raw hex / rgb / hsl / oklch outside designated token paths. Reports file:line with replacement suggestion. - `scripts/validate-metadata.py` — validate this skill's `name`/`description` frontmatter against the agentskills.io spec. ## Tunable Design Dials Three named dials let the user steer taste without dragging the conversation through dozens of micro-preferences. Declare values before generating anything visual, and pin them in the deliverable so a future review can re-anchor. | Dial | Range | Default | Low end (1–3) | High end (8–10) | | --------------------- | ----- | ------- | -------------------------------------------- | ----------------------------------------------------- | | `VISUAL_VARIANCE` | 1–10 | 6 | Conservative, system-aligned, recognizable | Bold, expressive, willing to diverge from the category default | | `MOTION_INTENSITY` | 1–10 | 4 | Reduced/utility only: state feedback | Rich orchestration, layered materials, premium feel | | `INFORMATION_DENSITY` | 1–10 | 5 | Spacious, calm, marketing-leaning | Dense, data-heavy, dashboard-leaning | What changes at each end: - **`VISUAL_VARIANCE`** — at low values, defaults to the closest equivalent of a familiar pattern (e.g. shadcn defaults, Material 3, Apple HIG). At high values, breaks the category template deliberately — bento over grid, asymmetric over centred, branded color over neutral. - **`MOTION_INTENSITY`** — drives durations (low → 80–150ms only; high → up to 500ms entrances, layered easing), and unlocks premium materials (blur, mask, clip-path) at ≥ 7. - **`INFORMATION_DENSITY`** — drives spacing scale ramps (low → space-6/8/10; high → space-1/2/3), row heights, line-height, and chart annotation richness. High density implies tabular-nums and keyboard-first interactions. When the user asks for "more bold" / "calmer" / "tighter" / "looser", translate to dial moves. Do not re-derive every other choice from scratch. Cite: dial pattern adapted from `taste-skill` (Leonxlnx, open-design ecosystem). ## HARD-GATE ``` Do NOT generate, modify, or approve any user-facing UI surface until: 1. The brand authorities have been READ (not skimmed): - DESIGN.md (root) if present - Design token files (e.g. packages/ui/src/tokens.css, theme.ts, tailwind.config.*) - Frontend conventions doc (e.g. web/CLAUDE.md, .cursorrules, .ai/rules.md) - Copy authority (e.g. COPY.md, content/style-guide.md) 2. The surface's JOB is stated in one sentence (what the user is here to do). 3. The complete STATE LIST for the surface is enumerated (default, hover, active, focus-visible, disabled, loading, empty, error, success — plus any domain states). 4. The component is known to either EXIST in the design system already or is JUSTIFIED as a new addition with a token-level proposal. 5. The SCENE SENTENCE has been written (see below). Categories are not scenes. 6. The REGISTER has been picked: Product or Brand (see below).
Related in Design
contribute
IncludedLocal-only OSS contribution command center. Auto-refreshes the user's in-flight PR and issue state on invoke so conversations start with full context — no need to brief Claude on what's in flight. Helps the user find issues to contribute to on GitHub, builds per-repo dossiers of what each upstream expects (CLA, DCO, branch convention, AI policy, draft-first, review bots, issue templates), runs deterministic gates before any external action so AI-assisted contributions don't reach maintainers as slop. State is markdown-only: candidate files at ~/.contribute-system/candidates/, repo dossiers at ~/.contribute-system/research/, append-only event log at ~/.contribute-system/log.jsonl. No database, no cloud calls. Use when the user asks about their PRs / issues / contributions, wants to find new work to take on, claim an issue, build/refresh a repo's dossier, or draft a Design Issue or PR. Trigger with "/contribute", "what's my PR status", "find a contribution", "claim issue X", "draft a Design Issue for Y", "refresh dossier for Z".
architectural-analysis
IncludedUser-triggered deep architectural analysis of a codebase or scoped subtree across eight modes — information architecture, data flow, integration points, UI surfaces, interaction patterns, data model, control flow, and failure modes. This skill should be used when the user asks to "diagram this codebase," "map the architecture," "show the data flow," "give me an ERD," "trace control flow," "find the integration points," "verify the layout pattern," "audit the UX architecture," or any similar request whose primary deliverable is mermaid diagrams plus cited reports under docs/architecture/. Dispatches haiku/sonnet sub-agents in parallel for per-mode exploration, then verifies every citation mechanically before any node lands in a diagram. Not for one-off prose explanations of code (use code-explanation) or for high-level system design from scratch (use system-design).
mcp
IncludedModel Context Protocol (MCP) server development and tool management. Languages: Python, TypeScript. Capabilities: build MCP servers, integrate external APIs, discover/execute MCP tools, manage multi-server configs, design agent-centric tools. Actions: create, build, integrate, discover, execute, configure MCP servers/tools. Keywords: MCP, Model Context Protocol, MCP server, MCP tool, stdio transport, SSE transport, tool discovery, resource provider, prompt template, external API integration, Gemini CLI MCP, Claude MCP, agent tools, tool execution, server config. Use when: building MCP servers, integrating external APIs as MCP tools, discovering available MCP tools, executing MCP capabilities, configuring multi-server setups, designing tools for AI agents.
react-native-skia
IncludedDesign, build, debug, and optimise high-polish animated graphics in React Native or Expo using @shopify/react-native-skia, Reanimated, and Gesture Handler. Use when the user wants canvas-driven UI, shaders, paths, rich text, image filters, sprite fields, Skottie, video frames, snapshots, web CanvasKit setup, or performance tuning for custom motion-heavy elements such as loaders, hero art, cards, charts, progress indicators, particle systems, or gesture-driven surfaces. Also use when the user asks for fluid, glow, glass, blob, parallax, 60fps/120fps, or GPU-friendly animated effects in React Native, even if they do not explicitly say "Skia". Do not use for ordinary form/layout work with standard views.
plaid
IncludedProduct Led AI Development — guides founders from idea to launched product. Six capabilities: Idea (discover a product idea), Validate (pressure-test the idea against fatal flaws, problem reality, competition, and 2-week MVP feasibility), Plan (vision intake + document generation), Design (translate image references into a design.md spec), Launch (go-to-market strategy), and Build (roadmap execution). Use when someone says "PLAID", "plaid idea", "help me find an idea", "product idea", "idea from my business", "idea from my expertise", "plaid validate", "validate my idea", "pressure-test", "is this idea good", "find fatal flaws", "validate the problem", "plan a product", "define my vision", "generate a PRD", "product strategy", "plaid design", "design from image", "translate image to design", "create design.md", "extract design tokens", "plaid launch", "go-to-market", "launch plan", "GTM strategy", "launch playbook", "plaid build", "build the app", "start building", or "execute the roadmap".
nextjs-framer-motion-animations
IncludedAdds production-safe Motion for React or Framer Motion animations to Next.js apps, including reveal, hover and tap micro-interactions, whileInView, stagger, AnimatePresence, layout and layoutId transitions, reorder, scroll-linked UI, and lightweight route-content transitions. Use when the user asks to add, refactor, or debug Motion or Framer Motion in App Router or Pages Router codebases, especially around server/client boundaries, reduced motion, LazyMotion, bundle size, hydration, or route transitions. Avoid for GSAP-style timelines, WebGL or 3D scenes, heavy scroll storytelling, or CSS-only effects unless Motion is explicitly requested.