kabirrgrover/mcp-server

MCP Server

A Model Context Protocol (MCP) server for integrating with LLMs like Claude through Poke.

Features

• MCP Protocol Support - Full implementation of the Model Context Protocol
• Tool Registration - Easy-to-use tool registration system
• FastAPI Integration - HTTP server for MCP communication
• Environment-based Configuration - Secure credential management

Quick Start

1. Install dependencies:

uv venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate
pip install -e .

Or using uv:

uv sync

2. Create a .env file:

cp .env.example .env
# Edit .env with your configuration

3. Run the server:

Option A: Using the quick start script (recommended):

python run_server.py

Option B: Using uvicorn directly:

uvicorn src.server:app --host 0.0.0.0 --port 3000

Option C: Using uv:

uv run uvicorn src.server:app --host 0.0.0.0 --port 3000

The server will run on http://localhost:3000/mcp by default.

Configuration

Create a .env file with the following variables:

# Server Configuration
PORT=3000
MCP_AUTH_TOKEN=your-secret-token-here  # Optional, for securing the endpoint

# Add your API credentials here
# API_KEY=your-api-key
# API_SECRET=your-api-secret

Using with Poke

To connect your MCP server with Poke, configure Poke to point to your server endpoint:

{
  "mcpServers": {
    "my-mcp-server": {
      "url": "http://localhost:3000/mcp",
      "authToken": "your-secret-token-here"
    }
  }
}

Project Structure

mcp-server/
├── src/
│   ├── __init__.py
│   ├── server.py          # Main server implementation
│   ├── tools/             # Tool implementations
│   │   ├── __init__.py
│   │   └── example.py     # Example tool
│   └── config.py          # Configuration management
├── pyproject.toml         # Project dependencies
├── .env.example           # Example environment variables
└── README.md              # This file

Adding New Tools

To add a new tool, create a file in src/tools/ and register it in src/server.py:

from mcp.server.models import Tool

async def my_tool_handler(arguments: dict) -> dict:
    # Your tool logic here
    return {"result": "success"}

# Register in server.py:
server.register_tool(
    Tool(
        name="my_tool",
        description="Description of what the tool does",
        inputSchema={
            "type": "object",
            "properties": {
                "param": {"type": "string", "description": "Parameter description"}
            }
        }
    ),
    my_tool_handler
)

Development

1. Install development dependencies:

uv sync --extra dev

2. Run tests:

pytest

3. Format code:

black src/
ruff check src/

License

MIT