Starlette MCP Sse

🚀 MCP Server-Sent Events (SSE) Implemented with Starlette Framework

This project implements Server-Sent Events (SSE) using the Starlette framework and integrates the Model Context Protocol (MCP).

🚀 Quick Start

This project demonstrates how to implement Server-Sent Events (SSE) in the Starlette framework while integrating the Model Context Protocol (MCP) functionality. Its core feature is to seamlessly combine the SSE functionality of MCP with a complete Starlette web application that includes custom routes.

✨ Features

• Server-Sent Events (SSE) implemented with MCP
• Integration with the Starlette framework, featuring custom routes
• A unified web application supporting both MCP and standard web endpoints
• Customizable routing structure
• Clear separation of responsibilities between MCP and web functionalities

📦 Installation

Prerequisites

Install the UV Package Manager - a fast Python package installer based on Rust:

powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"

Quick Run

Run the following command directly:

uvicorn main:app --reload

Then open your browser and visit http://localhost:8000.

Full Installation

1. Clone the repository:

   git clone https://github.com/your-repository.git
   cd your-repository
   

2. Create a virtual environment and install dependencies:

   python -m venv venv
   source venv/bin/activate  # Use `venv\Scripts\activate` on Windows
   pip install -r requirements.txt
   

3. Start the server:

   uvicorn main:app --reload
   

💻 Usage Examples

Available Endpoints

Visit the following URLs to test the functionality:

• / - Default home page
• /sieve?n=1000 - Example API for prime number screening
• /about - About page
• /docs - API documentation page

Debugging with MCP Inspector

To debug using the MCP Inspector, run the following command:

python -m mcp_inspector server --config inspector_config.json

Then open your browser and visit http://localhost:8091.

Extending the Application

Adding Custom Routes

Add new routes to the Starlette application, for example:

from fastapi import FastAPI
from starlette.routing import Route, Router

app = FastAPI()

async def root(request):
    return "Hello, World!"

async def sieve(request):
    n = int(request.query_params.get("n", 100))
    primes = []
    for num in range(2, n+1):
        if all(num % i != 0 for i in primes):
            primes.append(num)
    return {"primes": primes}

app.add_route("/", root)
app.add_route("/sieve", sieve, methods=["GET"])

Configuring the Continue Plugin

To use this MCP server in the Continue extension in VS Code, add the following configuration to your Continue settings:

{
  "experimental": {
    "modelContextProtocolServers": [
      {
        "name": "weather",
        "type": "sse",
        "url": "http://localhost:8000/sse"
      }
    ]
  }
}

This will allow you to directly use the functionality provided by this MCP server in Continue.

📚 Documentation

What is MCP?

The Model Context Protocol (MCP) is an open standard that allows AI models to interact with external tools and data sources. MCP addresses several key challenges in AI development:

• Context Limitation: Allows models to access the latest information beyond their training data
• Tool Integration: Provides a standardized way for models to use external tools and APIs
• Interoperability: Creates a common interface between different AI models and tools
• Scalability: Makes it easy to add new features to AI systems without retraining

This project demonstrates how to implement MCP using SSE in a Starlette web application.

Architecture

This project showcases a modular architecture that:

1. Integrates MCP SSE endpoints (/sse and /messages/) into the Starlette application
2. Provides standard web routes (/, /about, /status, /docs)
3. Demonstrates how to keep MCP functionality separate from web routes