home / mcp / gaggiuino mcp server

Gaggiuino MCP Server

Gaggiuino MCP server

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "andrewklement-gaggiuino-mcp": {
      "command": "uv",
      "args": [
        "--directory",
        "/ABSOLUTE/PATH/TO/PARENT/FOLDER/gaggiuino-mcp",
        "run",
        "gaggiuino.py"
      ]
    }
  }
}

This MCP server provides real-time access to Gaggiuino data, allowing clients to fetch the machine’s current status, the latest shot identifier, and detailed data for a given shot. It is designed to run on a local network and to be easily consumed by AI clients that visualize or analyze espresso data in real time.

How to use

After you have the Gaggiuino MCP server running, you can query it from your MCP client to access three core capabilities: check the machine status, retrieve the latest shot ID, and fetch data for a specific shot. Use the available toolsets to request status, get the latest shot identifier, and pull complete shot data for analysis or visualization. All interactions are designed to work over your local network with low latency for real-time analysis.

Typical usage flows involve first checking the machine’s status to ensure the device is ready, then obtaining the latest shot ID, and finally requesting data for that shot in order to analyze its telemetry and profile. You can chain these actions to build dashboards or automation that respond to new shots or changes in machine state.

How to install

Prerequisites: you need Node.js and npm or npx available on your system. If you plan to install tooling via npm or npx, ensure your environment has internet access to fetch packages.

Option A — Run the MCP server locally using the provided command snippet. This config demonstrates how to start the server through a local runtime (stdio-based) and points to the gaggiuino-mcp script.

{
  "mcpServers": {
      "gaggiuino": {
          "command": "uv",
          "args": [
              "--directory",
              "/ABSOLUTE/PATH/TO/PARENT/FOLDER/gaggiuino-mcp",
              "run",
              "gaggiuino.py"
          ]
      }
  }
}

Additional notes

Real-time access to shot telemetry is designed for local network usage, enabling you to build responsive dashboards or AI-assisted analyses without exposing data externally.

If you use a command-line tool to install or run the MCP server, you can leverage package managers like npx to automate the setup. For Claude Desktop integration, configure the stdio server entry as shown above and point your client to the appropriate local address.

Troubleshooting and tips

If the server does not respond, verify that the local machine can reach the gaggiuino-mcp directory and that gaggiuino.py is executable under the runtime used by uv.

Ensure the device running the server remains on the same local network as your MCP clients to minimize latency and avoid cross-network routing issues.

Available tools

getLatestShotId

Fetches the identifier of the most recently recorded espresso shot.

getShotData

Fetches detailed telemetry data for a specified shot identifier.

getStatus

Retrieves the current status of the espresso machine, including temperature, pressure, and flow characteristics.