MCP Server Manager V5r

🚀 MCP Server Guide

This guide provides a comprehensive overview of setting up, configuring, and using the MCP server, as well as integrating it with Cursor to enhance the functionality of your AI assistant.

🚀 Quick Start

✨ Features

• Install and run multiple servers easily.
• Configure servers through a JSON file.
• List, stop, and remove server configurations conveniently.
• Integrate with Cursor to expand AI assistant capabilities.

📦 Installation

Install Dependencies

1. Ensure you have Python and pip installed.
2. Install the necessary packages using the following command:

pip install psutil requests

Initialize Configuration

• Create and edit the mcp_config.json file to add your server configurations.

Start the MCP Server

1. Start all servers:

make run-servers

2. Start a single server:

python mcp_servers.py run <server_name>

💻 Usage Examples

List Servers

python mcp_servers.py list

Stop Servers

1. Stop all servers:

make stop-servers

2. Stop a single server:

python mcp_servers.py stop <server_name>

Remove Server Configuration

python mcp_servers.py remove <server_name>

📚 Documentation

Configuration Format

{
  "servers": [
    {
      "name": "github-api",
      "command": "python",
      "args": ["github_mcp_server.py"],
      "env": {
        "API_KEY": "${GITHUB_API_KEY}"
      },
      "port": 8080,
      "server_type": "stdio"
    },
    {
      "name": "perplexity-api",
      "command": "python",
      "args": ["perplexity_mcp_server.py"],
      "env": {},
      "port": null,
      "server_type": "sse"
    }
  ]
}

Integrate with Cursor

Configure Cursor's MCP Settings

1. Create or edit the ~/.cursor/mcp.json file.
2. Add your MCP servers, for example:

{
  "mcpServers": {
    "filesystem": {
      "url": "http://localhost:8090/sse"
    },
    "youtube": {
      "url": "http://localhost:8092/sse"
    }
  }
}

Available MCP Servers

• filesystem (Port 8090): Access the local file system.
• youtube (Port 8092): Search and retrieve YouTube videos.
• fetch (Port 8093): Perform HTTP requests.
• github (Port 8094): Interact with GitHub repositories.
• brave-search (Port 8096): Web search functionality.
• google-maps (Port 8098): Map and location services.

Test MCP Servers in Cursor

1. Open Cursor.
2. Start a new chat and enter questions like:
   • "List all files in my download folder."
   • "Search for 'AI tutorials' on YouTube."

By following these steps, you can easily configure and manage the MCP server and integrate it with Cursor to enhance the functionality of your AI assistant.