This repository provides a reusable template for a Python server implementing the Model Context Protocol (MCP), running in a Docker container and utilizing Streamable HTTP for real-time communication. Built on the Python implementation of the Model Context Protocol (MCP), this template enables easy integration with Large Language Models (LLMs).
The Model Context Protocol (MCP) lets you build servers that expose data and functionality to LLM applications in a secure, standardized way.
This template demonstrates a production-ready MCP server running in a Docker container, utilizing stateless Streamable HTTP for real-time communication.
dockerized-mcp-server-template/
├── src/
│ ├── Dockerfile
│ ├── requirements.txt
│ └── server.py
├── docker-compose.yml
└── README.md
docker-compose up --buildThe server will be accessible at:
http://localhost:8080/mcp
Alternatively, you can run the server directly using Python. First, install dependencies:
pip install -r src/requirements.txtThen run the server:
python src/server.pyThe server will be accessible at:
http://localhost:8080/mcp
The example includes a simple MCP tool function add:
@mcp.tool()
def add(a: int, b: int) -> int:
"""Add two numbers"""
return a + bYou can invoke this tool via MCP client requests.
This template uses Streamable HTTP transport, which is available in the Model Context Protocol starting with the Python SDK v1.8.0 (https://github.com/modelcontextprotocol/python-sdk).
Streamable HTTP significantly improves upon the previous SSE transport by enabling stateless operation, making MCP compatible with serverless architectures and standard web infrastructure. This approach eliminates the need for persistent connections, resulting in better resource efficiency and cost effectiveness, while offering a more straightforward implementation for developers. The result is an MCP server that's easier to deploy, more scalable, and more resource-efficient without sacrificing functionality.