Starwind UI MCP Server
Local MCP server implementation for Starwind UI that you can use with Cursor, Windsurf, and other AI tools - Boston343/starwind-ui-mcp
README
Starwind UI MCP Server
A TypeScript implementation of a Model Context Protocol (MCP) server for Starwind UI, providing tools to help developers work with Starwind UI components.
Quick Start
# Install dependencies
pnpm install
# Build the TypeScript code
pnpm build
# Start the server
pnpm start
Setup Your IDE
Instructions to set up your IDE to use a local MCP server vary by IDE. Here is an example for Windsurf:
{
"mcpServers": {
"starwind ui": {
"command": "node",
"args": ["c:\\path\\to\\folder\\starwind-ui-mcp\\dist\\server.js"],
"env": {}
}
}
}
Detailed instructions:
What is MCP?
The Model Context Protocol (MCP) is a protocol for extending AI capabilities through local servers. This implementation provides Starwind UI-specific tools to enhance AI assistant capabilities when working with Starwind UI. For more information about MCP itself, please visit the official documentation.
Features
- Tool-based Architecture - Modular design for easy addition of new tools
- Starwind UI Documentation Tool - Access documentation links for Starwind UI components
- Package Manager Detection - Detect and use the appropriate package manager (npm, yarn, pnpm)
- LLM Data Fetcher - Retrieve Starwind UI information for LLMs with caching and rate limiting
- TypeScript Implementation - Built with TypeScript for better type safety and developer experience
- Standard I/O Transport - Uses stdio for communication with AI assistants
Available Tools
Tool Name | Description |
---|---|
init_project |
Initializes a new Starwind UI project |
install_component |
Generates installation commands for Starwind UI components |
update_component |
Generates update commands for Starwind UI components |
get_documentation |
Returns documentation links for Starwind UI components and guides |
fetch_llm_data |
Fetches LLM data from starwind.dev (rate limited, with caching) |
get_package_manager |
Detects and returns the current package manager information |
Development
This project is set up to use PNPM for package manager for development purposes. If you are not using pnpm, you will need to update the package.json file with the appropriate package manager commands you need.
Project Structure
src/
├── config/ # Server configuration
│ └── settings.ts # Configuration settings
├── tools/ # MCP tools implementations
│ ├── index.ts # Tool registration
│ └── *.ts # Individual tool implementations
├── utils/ # Utility functions
└── server.ts # Main MCP server implementation
Adding New Tools
To add your own tools to the server:
- Create a new tool file in the
src/tools/
directory - Register the tool in
src/tools/index.ts
- Enable the tool in
src/config/settings.ts
- Rebuild the server with
pnpm build
- Restart the server with
pnpm start
License
MIT License - See LICENSE file for details.
推荐服务器

e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
AIO-MCP Server
🚀 集成了 AI 搜索、RAG 和多服务(GitLab/Jira/Confluence/YouTube)的一体化 MCP 服务器,旨在增强 AI 驱动的开发工作流程。来自 Folk。
https://github.com/Streen9/react-mcp
react-mcp 与 Claude Desktop 集成,能够根据用户提示创建和修改 React 应用程序。

any-chat-completions-mcp
将 Claude 与任何 OpenAI SDK 兼容的聊天完成 API 集成 - OpenAI、Perplexity、Groq、xAI、PyroPrompts 等。
Exa MCP Server
一个模型上下文协议服务器,它使像 Claude 这样的人工智能助手能够以安全和受控的方式,使用 Exa AI 搜索 API 执行实时网络搜索。
MySQL MCP Server
允许人工智能助手通过受控界面列出表格、读取数据和执行 SQL 查询,从而使数据库探索和分析更安全、更有条理。
browser-use MCP server
一个由人工智能驱动的浏览器自动化服务器,它实现了模型上下文协议,从而能够使用自然语言控制网页浏览器,以执行诸如导航、表单填写和视觉交互等任务。
mcp-codex-keeper
作为开发知识的守护者,为 AI 助手提供精心策划的最新文档和最佳实践访问权限。

OpenRouter MCP Server
提供与 OpenRouter.ai 的集成,允许通过统一的界面访问各种 AI 模型。
Supabase MCP Server
通过提供数据库管理、SQL查询执行和Supabase管理API访问工具,并内置安全控制,使Cursor和Windsurf能够安全地与Supabase数据库交互。