mcp-delonghi-ecam

MCP server for DeLonghi ECAM espresso coffee machines via local network control.

mcp-name: io.github.daedalus/mcp-delonghi-ecam

Overview

This MCP server provides comprehensive control over DeLonghi ECAM series automatic espresso coffee machines through their local network API. It enables LLMs to interact with the coffee machine for brewing beverages, monitoring status, and managing machine settings.

Supported Machines

• DeLonghi ECAM 550 series
• DeLonghi Dinamica Plus
• DeLonghi PrimaDonna Soul
• Other ECAM-based machines with local LAN API

Install

pip install mcp-delonghi-ecam

Usage

As MCP Server

Run the server:

mcp-delonghi-ecam

Or programmatically:

from mcp_delonghi_ecam import mcp

mcp.run()

Tools Available

Example Usage

# Connect to machine
await connect("192.168.1.100")

# Get status
status = await get_status("markdown")

# Brew espresso
result = await brew_beverage("espresso", 40, "normal", "medium")

# Brew cappuccino
result = await brew_beverage("cappuccino", 60, "strong", "high")

# Turn on cup warmer
await set_cup_warmer(True)

Configuration

The server connects to the coffee machine on the local network using:

• Default Port: 10280
• Protocol: HTTP (local LAN API)

Make sure the coffee machine is connected to the same network as the MCP server.

Development

git clone https://github.com/daedalus/mcp-delonghi-ecam.git
cd mcp-delonghi-ecam
pip install -e ".[test]"

# Run tests
pytest

# Format
ruff format src/ tests/

# Lint
ruff check src/ tests/

# Type check
mypy src/

API Reference

Beverage Types

• espresso, coffee, long_coffee, espresso_2x
• doppio_plus, americano, cappuccino
• latte_macchiato, caffe_latte, flat_white
• espresso_macchiato, hot_milk, hot_water, steam, chocolate

Taste Levels

• extra_mild, mild, normal, strong, extra_strong

Temperature Settings

• low, medium, high

License

MIT License - see LICENSE file for details.

Credits

Protocol information based on reverse-engineered specifications from:

• longshot (Rust implementation)
• ECAMpy
• delonghi-coffee-link-python