Model-Context-Protocol-MCP-Car-Control-Demo

harshalahah/Model-Context-Protocol-MCP-Car-Control-Demo

3.1

If you are the rightful owner of Model-Context-Protocol-MCP-Car-Control-Demo and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.

The Model Context Protocol (MCP) Car Control Demo is a learning project that demonstrates how to connect AI models to external systems using both a simple bridge and a real MCP server.

Model-Context-Protocol-MCP-Car-Control-Demo

A learning project exploring how to bridge AI with external systems. Includes two versions:

  1. A simple OpenAI API → UDP bridge
  2. A full MCP server tested with Claude Desktop.

What is MCP?

Model Context Protocol (MCP) is a new open standard (created by Anthropic, late 2024) that allows AI models to connect to external tools and data sources in a structured way. Think of it like a USB port for AI assistants -- instead of custom code for every tool, MCP provides a standard interface. AI can then call functions like 'unlock car' or 'check database' safely and consistently.

🚗 Model Context Protocol (MCP) Car Control Demo

This project is a learning showcase of how to connect AI models to external systems using both a simple bridge and a real MCP server.
The example: a fake car that can be locked or unlocked.

We built it in two stages:

🔑 Version 1 — OpenAI API → UDP Bridge

In this stage, we created:

  • A UDP Car Simulator — a Python program acting as a fake car listening for LOCK/UNLOCK commands.
  • A Flask HTTP → UDP Bridge — so that AI (or even curl) could call a simple HTTP endpoint, which then talks to the UDP car.

Test Commands

Run the UDP car simulator:

python udp_car_sim.py

In another terminal, unlock the car:

python -c "import socket; s=socket.socket(socket.AF_INET,socket.SOCK_DGRAM); s.sendto(b'UNLOCK',( '127.0.0.1',5005 )); print(s.recvfrom(1024)[0].decode())"

Run the bridge server:

python car_control_server.py

Then try:

curl http://127.0.0.1:8000/unlock_car

📝 We used Flask because it makes Python functions callable via HTTP without heavy setup.

🔑 Version 2 — Real MCP Server with Claude

In this stage, we upgraded to a proper MCP server using the fastmcp library. This let us connect directly with Claude Desktop (MCP-enabled), so the AI could issue LOCK/UNLOCK commands natively.

Files

  • server.py → the MCP server implementation
  • mcp.json → config file used by Claude Desktop

How it Works

  1. Claude asks the MCP server to run a command.
  2. The server sends UDP messages to the fake car.
  3. The car simulator responds back with success/failure.

⚠️ Notes

  • API Keys: If you experiment with OpenAI API, never hardcode your keys into code. Use environment variables.
  • This is for learning/demo purposes only (simulated car, not a real one).
  • MCP is a new standard — this project is meant as an introduction, not production code.

📚 What I Learned

  • Basics of TCP vs UDP networking
  • How to build a Flask bridge server
  • How MCP servers integrate with AI tools
  • Safe handling of API keys and configs

🚀 Next Steps

  • Expand to more car commands (start/stop, lights, etc.)
  • Try connecting other devices via MCP
  • Explore multi-client handling with threading

✨ This repo is part of my learning journey with AI, networking, and the new MCP ecosystem.