etudelab/agri-weather-mcp

Agri-Weather MCP

This project provides a comprehensive Weather Model Context Protocol (MCP) server specifically designed for agricultural applications. It leverages the free Open-Meteo API to provide real-time weather data, forecasts, and agricultural intelligence to support farming operations, cultivation planning, and crop management decisions.

Key Features

• Current Weather: Get real-time, localized weather conditions, including temperature, humidity, precipitation, and soil data.
• Weather Forecast: Access detailed 5-day weather forecasts with hourly granularity to support agricultural planning.
• Agricultural Alerts: Receive crop-specific alerts and recommendations for various growth stages (e.g., planting, vegetative, flowering).
• Soil Conditions: Retrieve detailed soil temperature and moisture information at multiple depths.
• Evapotranspiration Data: Obtain evapotranspiration (ET) and reference ET₀ data for precise irrigation management.
• Historical Weather: Access historical weather data for trend analysis and seasonal planning.

Technology Stack

• Python 3.11+
• MCP Server SDK: Core framework for building the Model Context Protocol server.
• HTTPX: For making asynchronous HTTP requests to the Open-Meteo API.
• Pytest: For running the comprehensive test suite.

Setup and Installation

1. Clone the repository:

   git clone <your-repository-url>
   cd agri-weather-mcp
   

2. Run the setup script (from repository root):

   ./scripts/setup.sh
   

Alternative Setup

1. Create and activate a Python virtual environment:

   python3 -m venv .venv
   source .venv/bin/activate
   

   On Windows, use .venv\Scripts\activate

2. Install the required dependencies:

   pip install -r requirements.txt
   

Usage

Running the Test Suite

The project includes a comprehensive test suite that verifies all functionality of the MCP server.

To run the tests, execute the following command from the project's root directory:

python -m pytest

This will run all tests using both the asyncio and trio backends to ensure compatibility with different async frameworks.

Test Features

• Comprehensive Coverage: Tests for all MCP tools and server components
• Multiple Region Support: Tests run against all supported regions (Indonesia, South East Asia, Australia, India)
• Mock API Responses: Uses realistic mock data that matches actual API response structures
• Async Testing: Compatible with both asyncio and trio backends
• Robust Error Handling: Tests for proper handling of API errors and invalid inputs

Test Structure

The test suite is organized as follows:

• test_mcp.py: Contains the main test classes and methods
• conftest.py: Contains pytest fixtures and mock data
• pytest.ini: Contains pytest configuration

The tests serve as practical examples of how to interact with the server's capabilities.

MCP Tools Overview

The server exposes the following tools through the Model Context Protocol:

1. get_current_weather: Provides real-time weather conditions for a specific location.

   • Parameters: latitude, longitude

2. get_weather_forecast: Delivers a 5-day weather forecast with daily and hourly data.

   • Parameters: latitude, longitude

3. get_agricultural_alerts: Generates crop-specific warnings and recommendations.

   • Parameters: latitude, longitude, crop_type, growth_stage

4. get_soil_conditions: Returns detailed soil temperature and moisture forecasts.

   • Parameters: latitude, longitude

5. get_evapotranspiration_data: Provides evapotranspiration data for irrigation planning.

   • Parameters: latitude, longitude

6. get_historical_weather: Fetches historical weather data for a specified date range.

   • Parameters: latitude, longitude, start_date, end_date

License

This project is licensed under the MIT License. See the file for details.

Attribution

This project uses weather data from Open-Meteo, which is licensed under the Creative Commons Attribution 4.0 International License (CC BY 4.0).

Weather data by Open-Meteo.com

When using this project, you must maintain this attribution wherever Open-Meteo data is displayed.