v0-mcp-ts
MCP server that bridges design and code by generating UI components, layouts, and themes from text prompts using v0.dev, with features like code refactoring and accessibility auditing.
README
v0-mcp-ts
The Ultimate Model Context Protocol (MCP) Server for AI-Powered Web Development, designed to bridge the gap between design and code.
v0-mcp-ts is a powerful, high-performance MCP server built with Bun and TypeScript. It integrates directly with v0.dev, a generative UI platform by Vercel, to bring AI-powered design and development to your workflow. Seamlessly generate UI from text prompts, iterate on designs like you would in Figma, and get production-ready code for modern web frameworks.
This server is your copilot for turning ideas into beautiful, functional, and accessible user interfaces.
<p align="center"> <img src="https://raw.githubusercontent.com/Conce-AI/v0.dev/main/assets/v0-mcp-ts-demo.png" alt="v0-mcp-ts Demo" width="800"> </p>
✨ Key Features
- 🚀 High-Performance: Built on Bun, providing a 25x faster development experience.
- 🎨 Design-to-Code: Leverages v0.dev to generate UI from text prompts and design iterations.
- 🧩 Component Generation: Create reusable components for React, Vue, Svelte, and more.
- 💅 Styling & Theming: Generate Tailwind CSS layouts and CSS-in-JS themes.
- ♿ Accessibility Audits: Ensure your components are WCAG compliant.
- 🔧 Code Refactoring: Improve existing components with AI-powered suggestions.
- ✅ Type-Safe: Written in TypeScript with strict type checking using Zod.
- 🧪 Comprehensive Testing: Includes a full suite of tests with Vitest.
🛠️ Tools Arsenal
This MCP server comes packed with a suite of tools to supercharge your development process:
| Tool | Description | Key Features |
|---|---|---|
generate_component |
Complete web component generation | React, Vue, Svelte support, TypeScript, Props |
generate_shadcn_component |
Create components using the shadcn/ui library | Leverages shadcn/ui primitives, variants, custom CSS |
generate_tailwind_layout |
Generate responsive layouts with Tailwind CSS | Sidebar, Header/Footer, Grid, Flex, Dark Mode |
generate_css_theme |
Design accessible CSS themes | WCAG AA compliance, CSS Vars, Tailwind Config |
refactor_component |
Refactor components with best practices | Performance, Accessibility, Typing, Modern Patterns |
audit_accessibility |
Comprehensive accessibility audit for your code | WCAG compliance, screen-reader, keyboard navigation |
💡 Prompts for Every Need
Use these prompts to guide the AI in generating exactly what you need:
| Prompt | Description | Use Case |
|---|---|---|
component-generator |
Generate React components with TypeScript | Create new UI components from scratch. |
shadcn-component-generator |
Generate components using shadcn/ui | Build complex UIs with a trusted component library. |
tailwind-layout-generator |
Generate responsive layouts with Tailwind CSS | Scaffold page layouts and responsive designs. |
css-theme-generator |
Generate accessible CSS themes | Create a consistent design system and color palette. |
component-refactor |
Refactor React components with best practices | Improve code quality, performance, and accessibility. |
accessibility-auditor |
Comprehensive accessibility audit tool | Ensure your application is usable by everyone. |
webapp-generator |
Generate complete web applications | Kickstart a new project with a solid foundation. |
🚀 Quick Start
1. Prerequisites
- Bun v1.2.0+
- v0.dev API Key
- An MCP Client (e.g., Cursor) =======
Installing via Smithery
To install v0-dev-mcp for Claude Desktop automatically via Smithery:
npx -y @smithery/cli install @nicotordev/v0-dev-mcp --client claude
2. Installation
# Clone the repository
git clone https://github.com/Conce-AI/v0.dev.git
cd v0-mcp-ts
# Install dependencies with Bun
bun install
# Run the interactive setup
bun run setup
3. Configuration
Configure your MCP client to connect to the server. Here's an example for Cursor:
// settings.json
{
"mcp.servers": {
"v0-mcp-ts": {
"command": "bun",
"args": ["run", "dev"],
"cwd": "/path/to/your/v0-mcp-ts",
"env": {
"V0_API_KEY": "your-v0-api-key"
}
}
}
}
🎬 Usage Examples
Generate a new Component
{
"tool": "generate_component",
"arguments": {
"component_name": "UserProfileCard",
"component_description": "A card to display user information with an avatar, name, and bio.",
"theme": "dark",
"styling_system": "tailwind"
}
}
Refactor an existing Component
{
"tool": "refactor_component",
"arguments": {
"code": "/* your existing component code here */",
"focus_areas": "performance,accessibility"
}
}
Audit for Accessibility
{
"tool": "audit_accessibility",
"arguments": {
"code": "/* your JSX/HTML code here */",
"audit_level": "wcag-aa"
}
}
🏗️ Supported Technologies
- Frontend Frameworks: Next.js, React, Vue, Svelte, Angular
- Styling: Tailwind CSS, CSS Modules, Styled Components, Emotion
- Testing: Vitest, Jest, Cypress, Playwright
🧪 Development & Testing
# Run development server with hot reload
bun run dev
# Run all tests
bun test
# Run tests with coverage report
bun run test:coverage
# Type-check the codebase
bun run type-check
🤝 Contributing
Contributions are welcome! Please see CONTRIBUTING.md for more details.
- Fork the repository.
- Create a new feature branch:
git checkout -b feature/your-amazing-feature - Commit your changes:
git commit -m 'Add some amazing feature' - Push to the branch:
git push origin feature/your-amazing-feature - Open a Pull Request.
📜 License
This project is licensed under the MIT License - see the LICENSE file for details.
👨💻 Author
Nicolas Torres
- 🌍 Location: Chile
- 🐙 GitHub: @nicotordev
- 💼 LinkedIn: nicotordev
- 🎥 YouTube: NicoTorDev Channel
A Full Stack Developer from Chile, passionate about modern web technologies, AI integration, and building high-performance applications.
🌟 Show Your Support
If this project helps you build amazing things, please give it a ⭐ on GitHub!
<p align="center"> <a href="https://github.com/sponsors/nicotordev"> <img src="https://img.shields.io/badge/Sponsor-GitHub-pink?style=for-the-badge&logo=github" alt="GitHub Sponsors"> </a> </p>
<p align="center"> <sub>Built with ❤️ and ⚡ Bun by <a href="https://github.com/nicotordev">Nicolas Torres</a> and Conce AI L.L.C. in Chile 🇨🇱</sub> </p>
Note on Legal Entities: Conce AI L.L.C. is a US-based company, while Conce AI spA is the Chilean entity.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。