open-design-ai
```markdown
What this skill does
```markdown
---
name: open-design-ai
description: Local-first open-source alternative to Claude Design — wire your existing coding agent (Claude Code, Codex, Cursor, Gemini CLI) into a skill-driven design workflow with 19 skills and 71 brand-grade design systems.
triggers:
- set up open design locally
- use open design with my coding agent
- generate a design artifact with open design
- add a new skill to open design
- pick a design system in open design
- export HTML or PDF from open design
- run the open design daemon
- configure open design with my API key
---
# open-design-ai
> Skill by [ara.so](https://ara.so) — Daily 2026 Skills collection.
Open Design (OD) is the open-source, local-first alternative to Anthropic's Claude Design. It turns your existing coding agent (Claude Code, Codex CLI, Cursor Agent, Gemini CLI, OpenCode, Qwen) into a design engine backed by **19 composable Skills** and **71 brand-grade Design Systems**. Artifacts render in a sandboxed iframe and export to HTML, PDF, PPTX, ZIP, or Markdown.
---
## Installation
### Prerequisites
- Node.js ≥ 18
- pnpm ≥ 8
- One supported coding agent on `PATH`: `claude`, `codex`, `cursor`, `gemini`, `opencode`, or `qwen`
### Three-command quickstart
```bash
git clone https://github.com/nexu-io/open-design.git
cd open-design
pnpm install
pnpm dev
```
Open `http://localhost:3000`.
### Vercel deploy
```bash
pnpm build
vercel deploy
```
### Single-process production
```bash
pnpm build
npm start
```
---
## Environment variables
Create `.env.local` (never commit this file):
```bash
# Required only when using the Anthropic API BYOK fallback
# (not needed if you rely on a local agent CLI)
ANTHROPIC_API_KEY=$ANTHROPIC_API_KEY
# Optional: pin which agent the daemon prefers
OD_AGENT=claude # claude | codex | cursor | gemini | opencode | qwen
# Optional: override the port the local daemon listens on
OD_DAEMON_PORT=4242
# Optional: working directory for on-disk project folders
OD_WORKSPACE_DIR=$HOME/.open-design/workspace
```
The daemon PATH-scans for agents automatically; `OD_AGENT` just sets priority.
---
## Architecture overview
```
Browser (Next.js)
│ streaming SSE / WebSocket
▼
Local Daemon (src/daemon/)
│ spawns agent CLI in a real on-disk project folder
▼
Agent CLI (claude / codex / gemini / …)
│ Read · Write · Bash · WebFetch against workspace
▼
Skill stack (skills/<name>/SKILL.md + seed template + checklist)
│ structured <artifact> tag in agent output
▼
Sandboxed iframe preview (srcdoc, vendored React 18 + Babel)
```
---
## Key concepts
### Skills
Each skill lives under `skills/<name>/` and follows the Claude Code `SKILL.md` convention with an extended `od:` frontmatter block:
```yaml
# skills/web-prototype/SKILL.md
---
name: web-prototype
description: Interactive single-page web prototype
od:
mode: prototype # prototype | deck | template
platform: web
scenario: landing
preview: iframe
design_system: linear # default design system slug
triggers:
- build me a landing page
- create a web prototype
- make a SaaS homepage
---
```
Skills are discovered at startup by scanning `skills/*/SKILL.md`.
### Design Systems
71 design systems live under `design-systems/<slug>/DESIGN.md`. Each exposes:
- A 4-colour OKLch signature palette
- Typography stack (font family + scale)
- Spacing and radius tokens
- A live `showcase.html`
Reference a system by slug anywhere in the UI or in a skill's `od.design_system` field.
### Visual Directions (no-brand fallback)
When the user has no existing brand the agent emits a direction-picker form with five curated schools:
| Slug | School |
|---|---|
| `editorial-monocle` | Editorial Monocle |
| `modern-minimal` | Modern Minimal |
| `tech-utility` | Tech Utility |
| `brutalist` | Brutalist |
| `soft-warm` | Soft Warm |
Each direction ships a deterministic OKLch palette + font stack — no model freestyle.
---
## Running a design session (programmatic)
### POST /api/design — start a session
```typescript
// src/app/api/design/route.ts (simplified excerpt)
const response = await fetch('/api/design', {
method: 'POST',
headers: { 'Content-Type': 'application/json' },
body: JSON.stringify({
skill: 'web-prototype', // skill slug
designSystem: 'linear', // design system slug (optional)
direction: 'modern-minimal', // visual direction (optional, no-brand fallback)
brief: 'A SaaS pricing page for a developer tool',
discovery: { // answers from the turn-1 discovery form
surface: 'web',
audience: 'developers',
tone: 'technical',
scale: 'single-page',
},
}),
});
// Response is a Server-Sent Events stream
const reader = response.body!.getReader();
const decoder = new TextDecoder();
while (true) {
const { done, value } = await reader.read();
if (done) break;
const chunk = decoder.decode(value);
// Parse SSE lines: data: {"type":"todo"|"artifact"|"tool_call"|"done", ...}
for (const line of chunk.split('\n')) {
if (line.startsWith('data: ')) {
const event = JSON.parse(line.slice(6));
handleEvent(event);
}
}
}
```
### Event types
```typescript
type DesignEvent =
| { type: 'todo'; todos: Todo[] } // live TodoWrite plan
| { type: 'tool_call'; tool: string; input: unknown }
| { type: 'artifact'; html: string } // final renderable artifact
| { type: 'done' }
| { type: 'error'; message: string };
```
---
## Working with Skills
### Add a custom skill
```bash
mkdir -p skills/my-skill
cat > skills/my-skill/SKILL.md << 'EOF'
---
name: my-skill
description: One-line description of what this skill produces
od:
mode: prototype
platform: web
scenario: custom
preview: iframe
design_system: vercel
triggers:
- build me a custom thing
- create a my-skill artifact
---
# My Skill
## What you produce
A single self-contained HTML file that …
## Checklist (P0 — must ship)
- [ ] Responsive at 375 px, 768 px, 1280 px
- [ ] No external image URLs (inline SVG or CSS gradients only)
- [ ] All colours from the active design system palette
EOF
```
OD hot-reloads skills in development — no restart needed.
### Add a seed template
Seed templates pre-populate the agent's workspace before generation:
```bash
mkdir -p skills/my-skill/seed
cat > skills/my-skill/seed/index.html << 'EOF'
<!DOCTYPE html>
<!-- SEED: agent reads this file first via the pre-flight Read step -->
<html lang="en">
<head><meta charset="UTF-8" /><title>{{title}}</title></head>
<body><!-- replace with generated content --></body>
</html>
EOF
```
---
## Working with Design Systems
### Browse available systems
```typescript
import { listDesignSystems } from '@/lib/design-systems';
const systems = await listDesignSystems();
// [{ slug: 'linear', name: 'Linear', palette: ['#5E6AD2', ...], ... }, ...]
```
### Load a specific system
```typescript
import { loadDesignSystem } from '@/lib/design-systems';
const ds = await loadDesignSystem('stripe');
console.log(ds.palette); // ['#635BFF', '#0A2540', '#00D924', '#FFFFFF']
console.log(ds.fonts); // { display: 'Sohne', body: 'Sohne', mono: 'Sohne Mono' }
console.log(ds.tokens); // spacing, radius, shadow tokens
```
### Add a new design system
```bash
mkdir -p design-systems/my-brand
cat > design-systems/my-brand/DESIGN.md << 'EOF'
---
name: My Brand
slug: my-brand
palette:
- '#1A1A2E' # primary
- '#16213E' # secondary
- '#0F3460' # accent
- '#E94560' # highlight
fonts:
display: 'Inter'
body: 'Inter'
mono: 'JetBrains Mono'
tokens:
radius: '8px'
spacing-unit: '8px'
---
## My Brand Design System
Use OKLch for all colour derivations. Primary background is near-black (`#1A1A2E`).
Accent `#E94560` for CTAs only — max one per viewport.
EOF
```
---
## Export
### From the UI
After an artifact renders, the export bar offers: **HTML · PDF · PPTX · ZIP · Markdown**.
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.