netatmo-java-mcp

kdubois/netatmo-java-mcp

3.2

If you are the rightful owner of netatmo-java-mcp and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to dayong@mcphub.com.

This server connects to the Netatmo Weather Station API to retrieve and expose weather data through REST and MCP interfaces.

Tools
3
Resources
0
Prompts
0

Netatmo Weather Station MCP Server (and optional JSON REST endpoint as well)

A Quarkus REST AND MCP Server that connects to the Netatmo Weather Station API and retrieves weather data.

Features

  • Model Context Protocol (MCP) tools for AI assistants
  • RESTful endpoints to retrieve weather station data
  • JSON response models for all Netatmo API data structures
  • Support for retrieving data from all stations or specific devices
  • Historical weather data with flexible parameters
  • OAuth2 authentication with Netatmo API using refresh tokens

Model Context Protocol (MCP) Server

This application exposes weather data as MCP tools that can be used by AI assistants and other MCP clients.

MCP Endpoints

  • Main MCP endpoint: http://localhost:8080/mcp (Streamable HTTP - 2025-03-26 protocol)
  • SSE endpoint: http://localhost:8080/mcp/sse (HTTP/SSE - 2024-11-05 protocol)

Available MCP Tools

get_current_weather

Gets current weather data from the Netatmo weather station.

  • Returns temperature, humidity, pressure, CO2, noise levels
  • Includes both indoor and outdoor measurements
  • Returns data in JSON format
get_available_devices

Lists all available Netatmo weather station devices.

  • Shows device IDs, station names, types, and supported data types
  • Useful for getting device IDs for historical data requests
  • Returns data in JSON format
get_historical_weather

Gets historical weather data from the Netatmo weather station.

  • Parameters:
    • deviceId (optional): Device ID, uses first device if not provided
    • scale (optional): Data granularity (30min, 1hour, 3hours, 1day, 1week, 1month), default: 1hour
    • sensorTypes (optional): Comma-separated sensor types (Temperature,Humidity,Pressure,CO2,Noise), default: Temperature,Humidity,Pressure
    • beginDate (optional): Begin date in format YYYY-MM-DD, default: 7 days ago
    • endDate (optional): End date in format YYYY-MM-DD, default: current date
    • maxDataPoints (optional): Maximum number of data points to return, default: all
  • Returns data in JSON format

Using MCP Tools

MCP tools can be used by:

  • AI assistants (Claude, GPT, etc.) that support MCP
  • MCP clients and inspectors
  • Custom applications using MCP protocol

Example MCP client usage:

# Connect to the MCP server
curl -X POST http://localhost:8080/mcp \
  -H "Content-Type: application/json" \
  -d '{"method": "tools/call", "params": {"name": "get_current_weather"}}'

REST Endpoints

Get All Weather Stations Data

GET /weather/stations

Returns complete weather station data including all devices and modules.

Get Current Weather Data

GET /weather/current

Returns simplified current weather data (temperature, humidity, pressure, CO2, noise) from the first station.

Get Available Devices

GET /weather/devices

Returns a list of available weather station devices with their IDs, names, types, and supported data types. Use this to get device IDs for historical data requests.

Get Historical Weather Data

GET /weather/historical

Returns historical weather data. If no device_id is provided, automatically uses the first available device. Supports query parameters:

  • device_id (optional): Specific device ID. If not provided, uses the first available device.
  • module_id (optional): Specific module ID
  • scale (optional): Data granularity (30min, 1hour, 3hours, 1day, 1week, 1month). Default: 1hour
  • type (optional): Sensor types (Temperature, Humidity, Pressure, CO2, Noise, Rain, WindStrength, WindAngle, GustStrength, GustAngle). Default: Temperature,Humidity,Pressure
  • date_begin (optional): Unix timestamp or ISO date (YYYY-MM-DD) for start date. Default: 7 days ago
  • date_end (optional): Unix timestamp or ISO date (YYYY-MM-DD) for end date. Default: now
  • limit (optional): Maximum number of data points. Default: 1024

Health Check

GET /weather

Returns a simple status message.

Configuration

The application is configured via environment variables:

# Netatmo API Configuration
NETATMO_API_BASE_URL=https://api.netatmo.com
NETATMO_API_CLIENT_ID=YOUR_CLIENT_ID
NETATMO_API_CLIENT_SECRET=YOUR_CLIENT_SECRET
NETATMO_API_REFRESH_TOKEN=YOUR_REFRESH_TOKEN

These are referenced in application.properties:

# Netatmo API Configuration
netatmo.api.base.url=${NETATMO_API_BASE_URL}
netatmo.api.client-id=${NETATMO_API_CLIENT_ID}
netatmo.api.client-secret=${NETATMO_API_CLIENT_SECRET}
netatmo.api.refresh-token=${NETATMO_API_REFRESH_TOKEN}

# REST Client Configuration
quarkus.rest-client.netatmo-api.url=${netatmo.api.base.url}
quarkus.rest-client.netatmo-api.scope=jakarta.inject.Singleton

# MCP Server Configuration
quarkus.mcp.server.traffic-logging.enabled=true
quarkus.mcp.server.traffic-logging.text-limit=1000
quarkus.mcp.server.sse.root-path=mcp

Setup Requirements

  1. Netatmo Developer Account: Create an application at Netatmo Connect

  2. OAuth2 Credentials: Obtain client_id, client_secret, and refresh_token

  3. Weather Station: Have a registered Netatmo weather station

  4. Environment Variables: Set the required environment variables. Recommended is to put them in a .env file, then you can use it to create a Kubernetes secret as such:

     kubectl create secret generic netatmo --from-env-file=./.env

Authentication Flow

This application uses OAuth2 refresh token authentication. The refresh token is used to automatically obtain access tokens for API calls without requiring user interaction.

Data Models

The application includes comprehensive DTOs for:

  • ApiResponse - Generic response wrapper for consistent formatting
  • NetatmoStationsDataResponse - Main API response wrapper
  • CurrentWeatherData - Simplified current weather data
  • HistoricalWeatherData - Historical weather measurements
  • DeviceInfo - Weather station device information
  • BaseResult - Base class for result objects

Example Response

{
  "success": true,
  "message": "Successfully retrieved current weather data",
  "data": {
    "temperature": 22.5,
    "humidity": 65,
    "pressure": 1013.2,
    "co2": 456,
    "noise": 42,
    "stationName": "Home Weather Station",
    "timeUtc": 1640995200
  },
  "status": 200
}

Running the Application

# Set environment variables
export NETATMO_API_BASE_URL=https://api.netatmo.com
export NETATMO_API_CLIENT_ID=your_client_id
export NETATMO_API_CLIENT_SECRET=your_client_secret
export NETATMO_API_REFRESH_TOKEN=your_refresh_token

# Development mode
./mvnw quarkus:dev

# Package and run
./mvnw package
java -jar target/quarkus-app/quarkus-run.jar

Container Support

The application includes container support with several Dockerfile options (in src/main/docker):

# Build using the Quarkus CLI:
quarkus image build

# Build using mvn and native compilation:
quarkus image build --native

# Build using docker, in JVM mode
docker build -f src/main/docker/Dockerfile.jvm -t quarkus/netatmo .

# Run the container
docker run -i --rm -p 8080:8080 \
  -e NETATMO_API_BASE_URL=https://api.netatmo.com \
  -e NETATMO_API_CLIENT_ID=your_client_id \
  -e NETATMO_API_CLIENT_SECRET=your_client_secret \
  -e NETATMO_API_REFRESH_TOKEN=your_refresh_token \
  quarkus/netatmo

Dependencies

  • Quarkus REST with Jackson
  • Quarkus OIDC Client for OAuth2 authentication
  • Quarkus REST Client for HTTP calls
  • Quarkus MCP Server for Model Context Protocol support