pyBittle-mcp-server

A Python MCP (Model Context Protocol) server for controlling the Bittle robot via Bluetooth. This server exposes a set of commands to move, pose, and interact with Bittle using the MCP protocol.

Features

• Connects to Bittle via Bluetooth

• Exposes movement and pose commands (forward, backward, turn, sit, rest, etc.)

• Integrates with MCP tools for remote and programmatic control

• Logs all activity to bittle_mcp.log

Related MCP server: Bluetooth MCP Server

Requirements

• Python 3.10+

• uv for package management

• Bittle robot with Bluetooth enabled

• pyBittle library

• mcp-server library

Installation & Setup

1. Install uv (recommended for fast, reliable Python package management):

   curl -Ls https://astral.sh/uv/install.sh | sh

2. Install uvx (required for MCP server):

   curl -Ls https://astral.sh/uvx/install.sh | sh

3. Install dependencies:

   # Option 1: Using start_server.sh (recommended) chmod +x start_server.sh ./start_server.sh # Option 2: Manual installation uv pip install -r requirements.txt # or, if using pyproject.toml: uv pip install -r pyproject.toml

Usage

1. Configure Bluetooth: Ensure your Bittle robot is powered on and in Bluetooth pairing mode.

2. Run the MCP server:

   # Option 1: Using start_server.sh (recommended) ./start_server.sh # Option 2: Manual start uvx mcpo --port 8080 -- uv run --with 'mcp[cli]' --with git+https://github.com/cluesang/pyBittle.git mcp run ./server.py

   The server will attempt to connect to Bittle and log status to bittle_mcp.log.

3. Integrate with MCP tools:

   • This server exposes commands via the MCP protocol, making them accessible to any MCP-compatible client or tool.

   • You can use the MCP CLI or other MCP tools to discover and invoke available commands on your Bittle robot.

   • Example (using MCP CLI):

     mcp call move_forward mcp call sit mcp call rest

   • All available commands are decorated with @mcp.tool() in server.py and are automatically registered with the MCP server.

Development

• The main logic is in server.py.

• Logging is configured to output to both console and bittle_mcp.log.

• Commands are decorated with @mcp.tool() for MCP exposure.

• Use start_server.sh for development as it handles environment setup and process management.

Troubleshooting

• If the server fails to connect, check Bluetooth pairing and ensure no other process is using the Bittle connection.

• Review bittle_mcp.log for detailed error messages.

• If you encounter port conflicts, start_server.sh will automatically handle killing existing server processes.

• Make sure both uv and uvx are installed and available in your PATH.

License

MIT License