mcp-meteoblue

mcp-meteoblue

Enables MCP clients to search for locations, fetch weather forecasts, and retrieve forecast meteogram images through the meteoblue Weather API.

Category
访问服务器

README

mcp-meteoblue

An MCP server for the meteoblue Weather API, written in JavaScript. It lets MCP clients search for locations, fetch forecasts using packages available through the Free Weather API, and request forecast meteogram images.

Tools

  • search_locations resolves cities, postal codes, and IATA/ICAO airport codes with meteoblue Location Search.
  • get_forecast accepts either a place name or coordinates and returns forecast JSON. It defaults to current, hourly, and daily weather.
  • get_forecast_image accepts either a place name or coordinates and returns a forecast image directly to the MCP client.

Place names passed to forecast or image tools are always resolved through the official meteoblue Location Search API. Forecast package inputs are restricted to the packages listed in the Free Weather API documentation.

get_forecast can restrict hourly output to a local time range. For example, tomorrow from 14:00 through 16:00 (inclusive):

{
  "location": "Paris",
  "forecastDays": 2,
  "dayOffset": 1,
  "startHour": "14:00",
  "endHour": "16:00"
}

You can use an explicit date such as 2026-08-24 instead of dayOffset. An end time earlier than the start time represents an overnight range. When a range is specified, the server requests basic-1h automatically and returns only the hourly data inside that range.

[!NOTE] meteoblue's current Free Weather API documentation says that images require a higher access level. The image tool is included for keys with an Image API entitlement and will return a clear authorization error otherwise.

Requirements

Keep the key private. This server reads it from METEOBLUE_API_KEY; it is never part of the package or MCP tool arguments.

Run from npm

Add this stdio server to your MCP client configuration:

{
  "mcpServers": {
    "meteoblue": {
      "command": "npx",
      "args": ["-y", "mcp-meteoblue"],
      "env": {
        "METEOBLUE_API_KEY": "your_api_key_here"
      }
    }
  }
}

Or run it directly:

METEOBLUE_API_KEY=your_api_key_here npx -y mcp-meteoblue

The server uses stdio, so it intentionally produces no normal terminal output while it waits for an MCP client.

Docker

Build the image locally:

docker build -t mcp-meteoblue .

Configure a Docker-based stdio server in your MCP client:

{
  "mcpServers": {
    "meteoblue": {
      "command": "docker",
      "args": [
        "run",
        "--rm",
        "-i",
        "--env",
        "METEOBLUE_API_KEY",
        "mcp-meteoblue"
      ],
      "env": {
        "METEOBLUE_API_KEY": "your_api_key_here"
      }
    }
  }
}

The container runs as the unprivileged node user. Keep stdin open with -i because MCP communication uses stdio.

Develop locally

npm install
npm test
METEOBLUE_API_KEY=your_api_key_here npm start

Point an MCP client at node /absolute/path/to/mcp-meteoblue/src/index.js and set the environment variable in that client's configuration.

Releases

GitHub Actions runs the test suite and validates the npm tarball on Node.js 20, 22, 24, and 26 for every push and pull request. Dependabot checks npm and GitHub Actions dependencies weekly.

Pushing a v* Git tag triggers .github/workflows/publish.yml. The workflow uses npm trusted publishing (OIDC) and automatically attaches provenance. Configure the npm trusted publisher with:

  • GitHub owner: unixfox
  • Repository: mcp-meteoblue
  • Workflow: publish.yml
  • Environment: npm

No npm token is stored in GitHub when trusted publishing is configured.

API scope

The server calls only:

  • https://www.meteoblue.com/{language}/server/search/query3
  • https://my.meteoblue.com/packages/{packages}
  • https://my.meteoblue.com/images/{forecast-image-type}

It does not expose meteoblue History, Dataset, Maps, Measurements, or other APIs.

License

MIT

推荐服务器

Baidu Map

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。

官方
精选
JavaScript
Playwright MCP Server

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。

官方
精选
TypeScript
Magic Component Platform (MCP)

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。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

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。

官方
精选
本地
TypeScript
VeyraX

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。

官方
精选
本地
graphlit-mcp-server

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。

官方
精选
TypeScript
Kagi MCP Server

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。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

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。

官方
精选