MCP Server with Docker

This project demonstrates how to integrate the Model Control Protocol (MCP) with OpenAI's API, enabling OpenAI to access and use tools exposed by an MCP server running in Docker.

Prerequisites

• Docker installed on your system

• Git (to clone the repository)

Related MCP server: OpenAI MCP Server

Project Structure

• server.py: The MCP server implementation with a tool

• client.py: A client that connects to the server and calls the agent

• Dockerfile: Instructions for building the Docker image

• requirements.txt: Python dependencies for the project

Data Flow Explanation

1. User Query: The user sends a query to the system (e.g., "What is our company's vacation policy?")

2. OpenAI API: OpenAI receives the query and available tools from the MCP server

3. Tool Selection: OpenAI decides which tools to use based on the query

4. MCP Client: The client receives OpenAI's tool call request and forwards it to the MCP server

5. MCP Server: The server executes the requested tool (e.g., retrieving knowledge base data)

6. Response Flow: The tool result flows back through the MCP client to OpenAI

7. Final Response: OpenAI generates a final response incorporating the tool data

Running with Docker

Step 1: Build the Docker image

docker build -t mcp-server .

Step 2: Run the Docker container

docker run -p 8050:8050 mcp-server

This will start the MCP server inside a Docker container and expose it on port 8050.

Running the Client

Once the server is running, you can run the client in a separate terminal:

python client.py

The client will connect to the server, list available tools, and call the agent to answer the query.

Troubleshooting

If you encounter connection issues:

1. Check if the server is running: Make sure the Docker container is running with docker ps.

2. Verify port mapping: Ensure the port is correctly mapped with docker ps or by checking the output of the docker run command.

3. Check server logs: View the server logs with docker logs <container_id> to see if there are any errors.

4. Host binding: The server is configured to bind to 0.0.0.0 instead of 127.0.0.1 to make it accessible from outside the container. If you're still having issues, you might need to check your firewall settings.

5. Network issues: If you're running Docker on a remote machine, make sure the port is accessible from your client machine.

Notes

• The server is configured to use SSE (Server-Sent Events) transport and listens on port 8050.

• The client connects to the server at http://localhost:8050/sse.

• Make sure the server is running before starting the client.