Caiyun Weather MCP Server

Model Context Protocol (MCP) is an open protocol that enables secure connections between AI applications and external data sources. The Caiyun Weather MCP server allows AI assistants (such as Claude) to directly access the Caiyun Weather API.

Connection Options

Before getting started, register and apply for a Caiyun Weather API key.

Remote Service

Connect directly to the Streamable HTTP MCP service hosted by Caiyun Weather:

{
  "url": "https://mcp-weather.caiyunapp.com/mcp",
  "headers": {
    "X-Caiyun-API-Key": "YOUR_CAIYUN_WEATHER_API_KEY"
  }
}

The outer configuration format depends on the MCP client. When calling a weather tool, the client must provide its Caiyun Weather API key in the X-Caiyun-API-Key request header.

Run Locally

Alternatively, run the public Python package locally with uvx:

{
  "mcpServers": {
    "caiyun-weather": {
      "command": "uvx",
      "args": ["mcp-caiyun-weather"],
      "env": {
        "CAIYUN_WEATHER_API_TOKEN": "YOUR_CAIYUN_WEATHER_API_KEY"
      }
    }
  }
}

Project repository: https://github.com/caiyunapp/mcp-caiyun-weather

Available Tools

The Caiyun Weather MCP server provides the following tools:

• get_realtime_weather: Get real-time weather data for a specific location (temperature, humidity, wind speed and direction, precipitation intensity, air quality, AQI, life indices, etc.)
• get_hourly_forecast: Get hourly weather forecast for the next 72 hours
• get_weekly_forecast: Get daily weather forecast for the next 7 days
• get_historical_weather: Get historical weather data for the past 24 hours
• get_weather_alerts: Get weather alerts for a specific location

For local installation, configuration, and development instructions, please refer to the GitHub repository.