open-meteo-mcp
MCP server for Open-Meteo free weather APIs, offering 16 tools for forecast, historical, air quality, marine, flood, ensemble, climate, seasonal, and scheduled data collection with no API key required.
README
open-meteo-mcp
MCP server for Open-Meteo — free weather APIs, no API key required.
Features
- 16 tools covering weather forecast, historical data, air quality, marine, flood, ensemble, climate projections, and scheduled data collection
- Scheduled jobs — server-side cron, persists readings to SQLite, returns aggregated stats
- No auth — Open-Meteo is free for non-commercial use
- stdio transport — works with Claude Desktop and any MCP client
Tools
Weather data
| Tool | Description |
|---|---|
geocode |
Search locations by name → lat/lon, timezone, country |
get_forecast |
7–16 day forecast, hourly + daily (45+ variables) |
get_historical |
ERA5 reanalysis from 1940 to present |
get_air_quality |
PM2.5, PM10, ozone, NO2, SO2, CO, pollen, EU/US AQI |
get_marine |
Wave height, period, direction, swell, sea surface temp |
get_elevation |
Elevation above sea level, batch up to 100 points |
get_flood |
GloFAS river discharge forecast, up to 16 weeks |
get_ensemble |
Multi-model ensemble with uncertainty quantification |
get_climate |
CMIP6 climate projections 1950–2050 |
get_seasonal |
ECMWF SEAS5 seasonal forecast up to 9 months |
get_dwd_icon |
DWD ICON model — high-res for Europe |
get_ecmwf |
ECMWF IFS model — gold standard global forecast |
Weather comparison
| Tool | Description |
|---|---|
compare_weather_cities |
Compare named cities on a date → ranked by weather quality score |
compare_weather_region |
Center city + radius km → grid sample → best spot in the area |
Score factors: sunshine, temperature comfort (18–28°C sweet spot), precipitation, wind, UV.
Example prompts:
- "Moscow, Sochi, Kazan this Saturday — where's best?" →
compare_weather_cities - "Best weather within 300 km of Moscow next Sunday?" →
compare_weather_region
Scheduled collection
| Tool | Description |
|---|---|
schedule_weather_job |
Create a recurring data collection job (cron expression) |
list_jobs |
List jobs with schedule, status, and reading count |
get_weather_summary |
Aggregated stats (min/max/avg) over a time window |
cancel_job |
Stop and delete a job |
Scheduled jobs survive server restarts — state is persisted in ~/.open-meteo-mcp/data.db.
Installation
git clone https://github.com/langepas/open-meteo-mcp.git
cd open-meteo-mcp
npm install
npm run build
Claude Desktop config
Add to ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"open-meteo": {
"command": "node",
"args": ["/absolute/path/to/open-meteo-mcp/dist/index.js"]
}
}
}
Restart Claude Desktop. Example prompts:
- "What's the weather in Moscow this week?" →
geocode→get_forecast - "Collect temperature in Saint Petersburg every hour" →
schedule_weather_job - "Show me the air quality summary for the last 24h" →
get_weather_summary
Development
npm run build # compile TypeScript
npm test # run unit tests
npm run dev # watch mode
Architecture
src/
├── index.ts # McpServer + StdioServerTransport entry point
├── client.ts # shared fetch helper (URLSearchParams, arrays → repeated keys)
├── db.ts # SQLite (better-sqlite3): jobs + readings tables
├── scheduler.ts # node-cron scheduler, restores jobs on start
└── tools/
├── forecast.ts # get_forecast
├── archive.ts # get_historical
├── air-quality.ts # get_air_quality
├── marine.ts # get_marine
├── elevation.ts # get_elevation
├── geocoding.ts # geocode
├── flood.ts # get_flood
├── ensemble.ts # get_ensemble
├── climate.ts # get_climate
├── seasonal.ts # get_seasonal
├── dwd-icon.ts # get_dwd_icon
├── ecmwf.ts # get_ecmwf
└── schedule.ts # schedule_weather_job, list_jobs, get_weather_summary, cancel_job
License
MIT
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。