matter

The matter-mcp-server exposes a structured MCP interface that lets Claude and other AI agents directly interact with Matter devices and protocol operations. It supports device commissioning and management, reading and writing node attributes, sending commands to Matter devices, monitoring events, and accessing Matter data models and documentation through the MCP protocol. You can discover and execute common tasks such as commissioning devices, querying node information, and sending commands from natural language prompts, while leveraging the included sample scripts to understand typical workflows. The server also integrates with Claude’s MCP tooling, enabling you to extend capabilities by adding additional MCP servers for related Matter utilities like datamodel browsing or protocol search.

To use the server, launch the Python script matter-mcp-server.py (via your preferred environment) and connect your MCP-enabled AI agent to the provided mcp endpoint. Explore the samples directory for practical examples such as starting event listeners, commissioning devices, querying node information, and sending device commands. You can also add extra MCP servers to the same configuration to broaden the agent's knowledge and capabilities about Matter protocol operations and data models.

Prerequisites:

• Python 3.8+ installed on your system
• A Python virtual environment tool (recommended: python3 -m venv)
• Access to Internet to install dependencies from requirements.txt
• Optional: uv tool for Claude integration (as described in the repository instructions)

Installation steps:

1. Clone the repository

git clone https://github.com/your-username/matter-mcp-server.git
cd matter-mcp-server

2. Create and activate a Python virtual environment

python3 -m venv .venv
source .venv/bin/activate  # On Windows use: .venv\Scripts\activate

3. Install dependencies

pip install -r requirements.txt

4. (Optional) Install uv for Claude integration per documentation

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

If curl is unavailable, you can use wget as an alternative:

wget -qO- https://astral.sh/uv/install.sh | sh

5. Configure Claude (if using Claude Desktop)

• Edit claude_desktop_config.json to include the matter-mcp-server entry as shown in the README.
• Restart Claude Desktop to load MCP tools.

6. Run the MCP server locally

python matter-mcp-server.py

Tips and common issues:

• Ensure your Python virtual environment is activated before installing dependencies and running the server.
• If you modify the configuration to add more MCP servers, ensure the command and arguments match the running environment (e.g., uv-based invocation for Claude integration).
• When using uv, replace placeholder paths with the actual full path to your repository or scripts.
• If you encounter connection issues with Claude, verify that the client is pointing to the correct MCP server name (e.g., matter-mcp-server) and that the server process is running.
• The samples directory contains ready-to-run scripts for common tasks like starting event listeners, commissioning devices, querying node information, and sending commands to devices.
• For testing without real devices, consider using a Matter Virtual Device (MVD) to validate basic interactions.