desAIgn is a high-performance AI designer that allows you to create websites, apps, and prototypes using natural language. Instead of generating raw code directly, it uses a unique Design Intent architecture that is token-efficient, predictable, and incredibly fast.
- Prompt-to-Design: Describe your vision in plain English and watch it come to life.
- Design Intent Architecture: Uses structured JSON "blueprints" for high reliability and low token usage.
- Instant Theme Swapping: Change colors and styles globally without re-generating code.
- Zero-Token Code Export: Deterministic client-side generation of clean React/Tailwind code.
- Interactive Canvas: A powerful visual workspace to view and iterate on your designs.
- Project Management: Save, share, and manage multiple design projects.
- Framework: Next.js 15 (App Router)
- Styling: Tailwind CSS 4
- AI Engine: Google Gemini AI via AI SDK
- Database: Neon (PostgreSQL) with Drizzle ORM
- Authentication: Better Auth
- Components: Custom UI library built on Radix UI primitives.
- Node.js (v18+ recommended)
- npm, pnpm, or yarn
- A Neon database instance (or any PostgreSQL)
- A Google AI Studio API Key for Gemini.
-
Clone the repository:
git clone https://github.com/your-repo/desaign.git cd desaign -
Install dependencies:
npm install
-
Set up environment variables: Create a
.envfile in the root directory and add the following:DATABASE_URL="your-postgresql-url" NEXT_PUBLIC_APP_URL="http://localhost:3000" BETTER_AUTH_SECRET="your-better-auth-secret" BETTER_AUTH_URL="http://localhost:3000" GEMINI_API_KEY="your-gemini-api-key" # Optional: For Auth Providers GITHUB_CLIENT_ID="your-github-id" GITHUB_CLIENT_SECRET="your-github-secret" GOOGLE_CLIENT_ID="your-google-id" GOOGLE_CLIENT_SECRET="your-google-secret"
-
Push the schema to your database:
npm run db:push
-
Seed initial themes and data:
npm run db:seed
Start the development server:
npm run devOpen http://localhost:3000 with your browser to see the result.
Desaign doesn't just "write code." It generates a Design Intent JSON structure. This approach has several advantages:
- Consistency: AI is better at following a JSON schema than writing complex JSX.
- Performance: Client-side rendering is faster than waiting for a full code generation.
- Flexibility: You can update the renderer independently of the AI's output.
For more technical details, check out DOCUMENTATION.md.
This project is licensed under the MIT License.