ui-design-system

$npx mdskill add alirezarezvani/claude-skills/ui-design-system

Generates design tokens and component libraries for UI designers

  • Automatically creates color palettes and typography scales.
  • Deploys tools to calculate responsive design breakpoints and spacing systems.
  • Uses machine learning algorithms to recommend optimal design token values.
  • Exports generated design tokens in various formats for integration into projects.

SKILL.md

.github/skills/ui-design-systemView on GitHub ↗
---
name: "ui-design-system"
description: UI design system toolkit for Senior UI Designer including design token generation, component documentation, responsive design calculations, and developer handoff tools. Use when creating design systems, generating design tokens, maintaining visual consistency, or facilitating design-dev collaboration and developer handoff.
---

# UI Design System

Generate design tokens, create color palettes, calculate typography scales, build component systems, and prepare developer handoff documentation.

---

## Table of Contents

- [Trigger Terms](#trigger-terms)
- [Workflows](#workflows)
  - [Workflow 1: Generate Design Tokens](#workflow-1-generate-design-tokens)
  - [Workflow 2: Create Component System](#workflow-2-create-component-system)
  - [Workflow 3: Responsive Design](#workflow-3-responsive-design)
  - [Workflow 4: Developer Handoff](#workflow-4-developer-handoff)
- [Tool Reference](#tool-reference)
- [Quick Reference Tables](#quick-reference-tables)
- [Knowledge Base](#knowledge-base)

---

## Trigger Terms

Use this skill when you need to:

- "generate design tokens"
- "create color palette"
- "build typography scale"
- "calculate spacing system"
- "create design system"
- "generate CSS variables"
- "export SCSS tokens"
- "set up component architecture"
- "document component library"
- "calculate responsive breakpoints"
- "prepare developer handoff"
- "convert brand color to palette"
- "check WCAG contrast"
- "build 8pt grid system"

---

## Workflows

### Workflow 1: Generate Design Tokens

**Situation:** You have a brand color and need a complete design token system.

**Steps:**

1. **Identify brand color and style**
   - Brand primary color (hex format)
   - Style preference: `modern` | `classic` | `playful`

2. **Generate tokens using script**
   ```bash
   python scripts/design_token_generator.py "#0066CC" modern json
   ```

3. **Review generated categories**
   - Colors: primary, secondary, neutral, semantic, surface
   - Typography: fontFamily, fontSize, fontWeight, lineHeight
   - Spacing: 8pt grid-based scale (0-64)
   - Borders: radius, width
   - Shadows: none through 2xl
   - Animation: duration, easing
   - Breakpoints: xs through 2xl

4. **Export in target format**
   ```bash
   # CSS custom properties
   python scripts/design_token_generator.py "#0066CC" modern css > design-tokens.css

   # SCSS variables
   python scripts/design_token_generator.py "#0066CC" modern scss > _design-tokens.scss

   # JSON for Figma/tooling
   python scripts/design_token_generator.py "#0066CC" modern json > design-tokens.json
   ```

5. **Validate accessibility**
   - Check color contrast meets WCAG AA (4.5:1 normal, 3:1 large text)
   - Verify semantic colors have contrast colors defined

---

### Workflow 2: Create Component System

**Situation:** You need to structure a component library using design tokens.

**Steps:**

1. **Define component hierarchy**
   - Atoms: Button, Input, Icon, Label, Badge
   - Molecules: FormField, SearchBar, Card, ListItem
   - Organisms: Header, Footer, DataTable, Modal
   - Templates: DashboardLayout, AuthLayout

2. **Map tokens to components**

   | Component | Tokens Used |
   |-----------|-------------|
   | Button | colors, sizing, borders, shadows, typography |
   | Input | colors, sizing, borders, spacing |
   | Card | colors, borders, shadows, spacing |
   | Modal | colors, shadows, spacing, z-index, animation |

3. **Define variant patterns**

   Size variants:
   ```
   sm: height 32px, paddingX 12px, fontSize 14px
   md: height 40px, paddingX 16px, fontSize 16px
   lg: height 48px, paddingX 20px, fontSize 18px
   ```

   Color variants:
   ```
   primary: background primary-500, text white
   secondary: background neutral-100, text neutral-900
   ghost: background transparent, text neutral-700
   ```

4. **Document component API**
   - Props interface with types
   - Variant options
   - State handling (hover, active, focus, disabled)
   - Accessibility requirements

5. **Reference:** See `references/component-architecture.md`

---

### Workflow 3: Responsive Design

**Situation:** You need breakpoints, fluid typography, or responsive spacing.

**Steps:**

1. **Define breakpoints**

   | Name | Width | Target |
   |------|-------|--------|
   | xs | 0 | Small phones |
   | sm | 480px | Large phones |
   | md | 640px | Tablets |
   | lg | 768px | Small laptops |
   | xl | 1024px | Desktops |
   | 2xl | 1280px | Large screens |

2. **Calculate fluid typography**

   Formula: `clamp(min, preferred, max)`

   ```css
   /* 16px to 24px between 320px and 1200px viewport */
   font-size: clamp(1rem, 0.5rem + 2vw, 1.5rem);
   ```

   Pre-calculated scales:
   ```css
   --fluid-h1: clamp(2rem, 1rem + 3.6vw, 4rem);
   --fluid-h2: clamp(1.75rem, 1rem + 2.3vw, 3rem);
   --fluid-h3: clamp(1.5rem, 1rem + 1.4vw, 2.25rem);
   --fluid-body: clamp(1rem, 0.95rem + 0.2vw, 1.125rem);
   ```

3. **Set up responsive spacing**

   | Token | Mobile | Tablet | Desktop |
   |-------|--------|--------|---------|
   | --space-md | 12px | 16px | 16px |
   | --space-lg | 16px | 24px | 32px |
   | --space-xl | 24px | 32px | 48px |
   | --space-section | 48px | 80px | 120px |

4. **Reference:** See `references/responsive-calculations.md`

---

### Workflow 4: Developer Handoff

**Situation:** You need to hand off design tokens to development team.

**Steps:**

1. **Export tokens in required formats**
   ```bash
   # For CSS projects
   python scripts/design_token_generator.py "#0066CC" modern css

   # For SCSS projects
   python scripts/design_token_generator.py "#0066CC" modern scss

   # For JavaScript/TypeScript
   python scripts/design_token_generator.py "#0066CC" modern json
   ```

2. **Prepare framework integration**

   **React + CSS Variables:**
   ```tsx
   import './design-tokens.css';

   <button className="btn btn-primary">Click</button>
   ```

   **Tailwind Config:**
   ```javascript
   const tokens = require('./design-tokens.json');

   module.exports = {
     theme: {
       colors: tokens.colors,
       fontFamily: tokens.typography.fontFamily
     }
   };
   ```

   **styled-components:**
   ```typescript
   import tokens from './design-tokens.json';

   const Button = styled.button`
     background: ${tokens.colors.primary['500']};
     padding: ${tokens.spacing['2']} ${tokens.spacing['4']};
   `;
   ```

3. **Sync with Figma**
   - Install Tokens Studio plugin
   - Import design-tokens.json
   - Tokens sync automatically with Figma styles

4. **Handoff checklist**
   - [ ] Token files added to project
   - [ ] Build pipeline configured
   - [ ] Theme/CSS variables imported
   - [ ] Component library aligned
   - [ ] Documentation generated

5. **Reference:** See `references/developer-handoff.md`

---

## Tool Reference

### design_token_generator.py

Generates complete design token system from brand color.

| Argument | Values | Default | Description |
|----------|--------|---------|-------------|
| brand_color | Hex color | #0066CC | Primary brand color |
| style | modern, classic, playful | modern | Design style preset |
| format | json, css, scss, summary | json | Output format |

**Examples:**

```bash
# Generate JSON tokens (default)
python scripts/design_token_generator.py "#0066CC"

# Classic style with CSS output
python scripts/design_token_generator.py "#8B4513" classic css

# Playful style summary view
python scripts/design_token_generator.py "#FF6B6B" playful summary
```

**Output Categories:**

| Category | Description | Key Values |
|----------|-------------|------------|
| colors | Color palettes | primary, secondary, neutral, semantic, surface |
| typography | Font system | fontFamily, fontSize, fontWeight, lineHeight |
| spacing | 8pt grid | 0-64 scale, semantic (xs-3xl) |
| sizing | Component sizes | container, button, input, icon |
| borders | Border values | radius (per style), width |
| shadows | Shadow styles | none through 2xl, inner |
| animation | Motion tokens | duration, easing, keyframes |
| breakpoints | Responsive | xs, sm, md, lg, xl, 2xl |
| z-index | Layer system | base through notification |

---

## Quick Reference Tables

### Color Scale Generation

| Step | Brightness | Saturation | Use Case |
|------|------------|------------|----------|
| 50 | 95% fixed | 30% | Subtle backgrounds |
| 100 | 95% fixed | 38% | Light backgrounds |
| 200 | 95% fixed | 46% | Hover states |
| 300 | 95% fixed | 54% | Borders |
| 400 | 95% fixed | 62% | Disabled states |
| 500 | Original | 70% | Base/default color |
| 600 | Original × 0.8 | 78% | Hover (dark) |
| 700 | Original × 0.6 | 86% | Active states |
| 800 | Original × 0.4 | 94% | Text |
| 900 | Original × 0.2 | 100% | Headings |

### Typography Scale (1.25x Ratio)

| Size | Value | Calculation |
|------|-------|-------------|
| xs | 10px | 16 ÷ 1.25² |
| sm | 13px | 16 ÷ 1.25¹ |
| base | 16px | Base |
| lg | 20px | 16 × 1.25¹ |
| xl | 25px | 16 × 1.25² |
| 2xl | 31px | 16 × 1.25³ |
| 3xl | 39px | 16 × 1.25⁴ |
| 4xl | 49px | 16 × 1.25⁵ |
| 5xl | 61px | 16 × 1.25⁶ |

### WCAG Contrast Requirements

| Level | Normal Text | Large Text |
|-------|-------------|------------|
| AA | 4.5:1 | 3:1 |
| AAA | 7:1 | 4.5:1 |

Large text: ≥18pt regular or ≥14pt bold

### Style Presets

| Aspect | Modern | Classic | Playful |
|--------|--------|---------|---------|
| Font Sans | Inter | Helvetica | Poppins |
| Font Mono | Fira Code | Courier | Source Code Pro |
| Radius Default | 8px | 4px | 16px |
| Shadows | Layered, subtle | Single layer | Soft, pronounced |

---

## Knowledge Base

Detailed reference guides in `references/`:

| File | Content |
|------|---------|
| `token-generation.md` | Color algorithms, HSV space, WCAG contrast, type scales |
| `component-architecture.md` | Atomic design, naming conventions, props patterns |
| `responsive-calculations.md` | Breakpoints, fluid typography, grid systems |
| `developer-handoff.md` | Export formats, framework setup, Figma sync |

---

## Validation Checklist

### Token Generation
- [ ] Brand color provided in hex format
- [ ] Style matches project requirements
- [ ] All token categories generated
- [ ] Semantic colors include contrast values

### Component System
- [ ] All sizes implemented (sm, md, lg)
- [ ] All variants implemented (primary, secondary, ghost)
- [ ] All states working (hover, active, focus, disabled)
- [ ] Uses only design tokens (no hardcoded values)

### Accessibility
- [ ] Color contrast meets WCAG AA
- [ ] Focus indicators visible
- [ ] Touch targets ≥ 44×44px
- [ ] Semantic HTML elements used

### Developer Handoff
- [ ] Tokens exported in required format
- [ ] Framework integration documented
- [ ] Design tool synced
- [ ] Component documentation complete

More from alirezarezvani/claude-skills

SkillDescription
a11y-auditAccessibility audit skill for scanning, fixing, and verifying WCAG 2.2 Level A and AA compliance across React, Next.js, Vue, Angular, Svelte, and plain HTML codebases. Use when auditing accessibility, fixing a11y violations, checking color contrast, generating compliance reports, or integrating accessibility checks into CI/CD pipelines.
ab-test-setupWhen the user wants to plan, design, or implement an A/B test or experiment. Also use when the user mentions "A/B test," "split test," "experiment," "test this change," "variant copy," "multivariate test," "hypothesis," "conversion experiment," "statistical significance," or "test this." For tracking implementation, see analytics-tracking.
ad-creativeWhen the user needs to generate, iterate, or scale ad creative for paid advertising. Use when they say 'write ad copy,' 'generate headlines,' 'create ad variations,' 'bulk creative,' 'iterate on ads,' 'ad copy validation,' 'RSA headlines,' 'Meta ad copy,' 'LinkedIn ad,' or 'creative testing.' This is pure creative production — distinct from paid-ads (campaign strategy). Use ad-creative when you need the copy, not the campaign plan.
adversarial-reviewerAdversarial code review that breaks the self-review monoculture. Use when you want a genuinely critical review of recent changes, before merging a PR, or when you suspect Claude is being too agreeable about code quality. Forces perspective shifts through hostile reviewer personas that catch blind spots the author's mental model shares with the reviewer.
aeoAnswer Engine Optimization (AEO) skill — optimize content to be cited by AI language models (ChatGPT, Perplexity, Claude, Gemini, Mistral) as authoritative sources. Distinct from SEO — AEO optimizes for citation in LLM-generated responses, not search rankings. Use when planning content for AI-first search audiences, auditing existing content for E-E-A-T signals, tracking which pages get cited by which LLMs, or building a citation-friendly content strategy. Triggers — 'AEO audit', 'optimize for ChatGPT', 'get cited by Perplexity', 'LLM citation strategy', 'answer engine optimization', 'content for AI search', 'E-E-A-T audit'. Output is a markdown audit report (default) or JSON for pipeline integration. Stdlib-only Python tools.
agent-decision-receiptsMint a tamper-evident, post-quantum-signed receipt for a consequential agent action (deploy, delete, pay, grant-access, model decision) so it can be verified later from the certificate alone. Use when an autonomous agent takes a side-effecting action that may need to be proven later, or when satisfying EU AI Act Article 12 record-keeping. Three decisions: whether an action needs a receipt, minting it, verifying it. Signing is delegated to the open-source OpenAgentOntology package. Not after-the-fact log analysis; not a hosted notary; not a legal opinion.
agent-designerUse when the user asks to design a multi-agent system, pick an orchestration pattern (supervisor/swarm/pipeline), generate tool schemas for agents, or evaluate agent execution logs for cost, latency, and failure bottlenecks. Examples: 'design an agent architecture for research automation', 'generate Anthropic tool schemas from these tool descriptions', 'analyze these agent run logs for bottlenecks'. NOT for Claude Code workflow files (use workflow-builder) or single-agent prompt design (use agent-workflow-designer).
agent-harnessTurn any domain folder of skills into a bounded agentic loop: compile a goal into a verifiable task plan, execute tasks with the domain's own tools, verify every task with machine-run checks, retry with caps, escalate to a human when budgets exhaust, and refuse to close until everything is verified or explicitly waived. Use when you want an agent or subagent to pick up a goal and drive it to a verified close across one of this repo's 18 domains ('run this goal through the engineering harness', 'set up an agentic loop for marketing work', 'make the finance domain self-verifying'). NOT for authoring Claude Code Workflow-tool .js scripts (workflow-builder), N-agent tournaments on one task (agenthub), single-file metric optimization (autoresearch-agent), or discovering published loop recipes (loop-library).
agent-launcher-orchestratorUse when a user wants to build, launch, grade, or schedule a Claude Managed Agent (CMA) in their own Anthropic account — "build me an agent", "launch this as a managed agent", "run this on a schedule", "grade my agent against a rubric", "set up a nightly worker". Reads the per-session goal (./my-agent/goal.json), routes deterministically to one of five phase sub-skills (interview → stage-launch → grade-iterate → run-without-you → wrap-up) via goal_router.py, and compiles the goal+phase into an execution shape (single-pass workflow / bounded grade→iterate loop / recurring cron deployment loop) via loop_compiler.py. Forks context so heavy intake (build sheets, payloads, eval cases) stays out of the parent thread. All launches are emitted as BYOK curl the user runs with their own key; no tool makes API calls. Inspired by anthropics/launch-your-agent (Apache-2.0). Distinct from engineering/agent-harness (generic domain loop) and engineering/write-a-skill (authors Claude Code skills, not CMAs).
agent-memoryUse when a project's CLAUDE.md has grown past what anyone reads and you want the agent to learn durable facts from its own sessions instead — or when asking why the agent keeps re-learning the same correction, why a remembered rule is wrong, or where a memory line came from. Implements a four-tier store (L0 transcripts / L1 candidates / L2 project context / L3 stable persona) where promotion is earned by recurrence across sessions and days, never by one confident statement, and nothing reaches a committed file without a human adopting it.