My MCP Server

A basic MCP (Model Context Protocol) server template demonstrating core capabilities.

Features

Tools

• echo - Echoes back a message (useful for testing)
• calculate - Performs basic math operations (add, subtract, multiply, divide)
• get_info - Returns information about the server

Resources

• resource://examples/welcome - Welcome message for new users
• resource://examples/capabilities - JSON description of server capabilities

Installation

1. Install dependencies:

pip install -r requirements.txt

2. Test the server locally:

python my_serv.py

Configuration

Add to Cursor

Add this server to your Cursor mcp.json file:

{
  "mcpServers": {
    "my-mcp-server": {
      "command": "python",
      "args": ["/Users/fpj/Development/python/mcp/my_serv.py"],
      "env": {
        "LOG_LEVEL": "INFO"
      }
    }
  }
}

Configuration Options

You can add environment variables to customize behavior:

"env": {
  "LOG_LEVEL": "DEBUG",
  "CUSTOM_VAR": "value"
}

Development

Project Structure

mcp/
├── my_serv.py          # Main server implementation
├── requirements.txt    # Python dependencies
└── README.md          # This file

Adding New Tools

To add a new tool:

1. Add the tool definition to list_tools():

Tool(
    name="my_tool",
    description="Description of what it does",
    inputSchema={
        "type": "object",
        "properties": {
            "param": {"type": "string"}
        },
        "required": ["param"]
    }
)

2. Add the tool handler in call_tool():

elif name == "my_tool":
    param = arguments.get("param")
    # Your logic here
    return [TextContent(type="text", text=f"Result: {param}")]

Adding Resources

To add a new resource:

1. Add to list_resources():

Resource(
    uri="resource://my/resource",
    name="Resource Name",
    mimeType="text/plain",
    description="What this resource provides"
)

2. Add the handler in read_resource():

elif uri == "resource://my/resource":
    return "Resource content here"

Usage in Cursor

After adding the server to your mcp.json and restarting Cursor:

1. Open any file in Cursor
2. In the chat, you can now use commands like:
   • "Use the echo tool to say hello"
   • "Calculate 15 + 27"
   • "Show me the server capabilities"

The AI will automatically call your MCP server tools!

Debugging

View Logs

Logs are written to stderr. In Cursor:

1. Go to View → Output
2. Select your server from the dropdown
3. View real-time logs

Common Issues

Server not starting:

• Check that Python 3.8+ is installed
• Verify dependencies are installed: pip install -r requirements.txt
• Check the path in mcp.json is correct

Tools not available:

• Restart Cursor (Cmd/Ctrl + Shift + P → "Developer: Reload Window")
• Check Output panel for error messages
• Verify mcp.json syntax is valid

Errors during tool execution:

• Check the Output panel for stack traces
• Ensure your tool logic handles edge cases
• Validate input parameters match the schema

Next Steps

Ideas to Extend This Server

1. File System Operations

   • Read/write files
   • Search directories
   • File management

2. API Integrations

   • Connect to external services
   • Fetch data from APIs
   • Send notifications

3. Database Access

   • Query databases
   • Execute SQL
   • Data analysis

4. Custom Workflows

   • Automate repetitive tasks
   • Code generation
   • Project scaffolding

5. System Operations

   • Run shell commands
   • Process monitoring
   • Environment management

Resources

• MCP Documentation
• MCP Python SDK
• Example MCP Servers

License

MIT