mcpsq
A simple MCP server to discover and manage MCP servers. Provides tools to find, add, list, and remove MCP server configurations for Cursor.
README
MCP^2(MCPSQ)
MCP^2(MCPSQ) is a simple MCP server to discover and manage MCP servers. See our website mcpsq.xyz for more information.
Usage
npm start
Local Development
Prerequisites
Installation
- Clone the repository:
git clone https://github.com/xinbenlv/mcpsq.git
cd mcpsq
- Install dependencies:
npm install
Running the Server
Start the MCP server locally:
# Production mode
npm start
# Development mode with auto-reload
npm run dev
# Build TypeScript to JavaScript
npm run build
The server will start on http://localhost:3000 and provide:
- SSE endpoint at
GET /mcp - Message handling at
POST /messages
Configuring Cursor MCP Integration
To use this MCP server with Cursor, add the following configuration to your Cursor MCP settings:
Option 1: Global Configuration
Create or edit ~/.cursor/mcp.json:
{
"mcpServers": {
"mcpsq": {
"url": "http://localhost:3000/mcp"
}
}
}
Option 2: Project-specific Configuration
Create .cursor/mcp.json in your project root:
{
"mcpServers": {
"mcpsq": {
"url": "http://localhost:3000/mcp"
}
}
}
Available MCP Tools
Once configured, the following tools will be available in Cursor:
find-mcp-server: Search for MCP servers relevant to your queryadd-mcp-server-to-cursor: Add an MCP server to your Cursor configurationlist-current-mcp-servers: List all currently configured MCP serversremove-mcp-server-from-cursor: Remove an MCP server from your configuration
Testing the Integration
- Start the MCPSQ server locally
- Configure Cursor with the MCP settings above
- Restart Cursor or reload the MCP configuration
- In Cursor chat, try asking: "Find MCP servers for database management"
- The MCPSQ tools should appear in the available tools list
Development Scripts
# Start the server in production mode
npm start
# Start the server in development mode with auto-reload
npm run dev
# Build TypeScript to JavaScript
npm run build
# Clean build directory
npm run clean
# Validate data files
npm run validate:data
# Type checking without emitting files
npm run typecheck
Development Workflow
- Development: Use
npm run devfor development with automatic restart on file changes - Type Checking: Run
npm run typecheckto check for TypeScript errors without building - Building: Use
npm run buildto compile TypeScript to JavaScript (automatically cleans first) - Production: Use
npm startto run the compiled TypeScript directly with tsx
Troubleshooting
- Server not starting: Check if port 3000 is available
- Cursor not connecting: Verify the MCP configuration file path and JSON syntax
- Tools not appearing: Check Cursor's MCP logs in the output panel
- TypeScript errors: Run
npm run typecheckto see detailed error messages
Heroku Deployment
MCPSQ can be easily deployed to Heroku for cloud hosting:
Prerequisites
- Heroku CLI installed
- Heroku account
Deployment Steps
- Login to Heroku:
heroku login
- Create a new Heroku app:
heroku create your-mcpsq-app-name
- Deploy to Heroku:
git push heroku main
- Open your deployed app:
heroku open
Configuration
The app will automatically:
- Use Heroku's
PORTenvironment variable - Build the TypeScript code during deployment via
heroku-postbuildscript - Start the server using the compiled JavaScript
Your MCPSQ server will be available at:
- MCP endpoint:
https://your-app-name.herokuapp.com/mcp
Using with Cursor
Configure Cursor to use your Heroku-deployed MCPSQ server:
{
"mcpServers": {
"mcpsq": {
"url": "https://your-app-name.herokuapp.com/mcp"
}
}
}
Data
The data is stored in the data directory.
Contributing
Contributions are welcome! Please feel free to submit a pull request that updates the data in the data directory.
ROADMAP
- [x] handle env vars needed
- [x] publish to npmjs.com
- [x] add Heroku deployment support
License
This project is licensed under the MIT License - see the LICENSE file for details.
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。