Weather

Weather is an MCP server that connects AI assistants like Claude Desktop and Cursor to real-time United States meteorological data and Feishu cloud documents. Built with pure JavaScript ES Modules, the server interfaces directly with the public US National Weather Service API without requiring external weather API credentials, providing instant access to regional forecasts and active weather alerts. In addition to meteorological queries, it integrates with the Feishu open platform, allowing AI agents to retrieve content across documents, spreadsheets, slides, multidimensional tables, and wiki entries. Developers, enterprise teams, and individual users employ this server to automate regional environmental monitoring alongside corporate knowledge retrieval within a single client interface. The server supports dual transport options, running over standard input and output (stdio) for local desktop applications or over HTTP Server-Sent Events (SSE) for remote and web deployments. Parameters are validated using Zod to ensure reliable execution when models query coordinates, state codes, or document identifiers.

Category: Maps, Weather & Local Data

Tags: forecast, meteorology, nws, weather

Visit Weather

How to install and configure Weather

1. Clone the repository and install dependencies by running npm install in the project root directory. 2. Configure Feishu credentials if using document tools by copying the template with cp .env.example .env and setting FEISHU_APP_ID and FEISHU_APP_SECRET. Weather features require no API keys. 3. For local clients like Claude Desktop or Cursor, configure the MCP client to execute node index.mjs stdio or npm start. 4. Alternatively, run the server in HTTP/SSE mode for web clients by executing node index.mjs sse or specifying a custom port with node index.mjs sse 3001.

What you can do with Weather

• Checking active weather alerts across specific US states using standard two-letter state codes like CA or NY. - Fetching point forecasts for precise coordinates within the United States using latitude and longitude inputs. - Retrieving plain text content from Feishu doc and docx cloud files using document URLs or IDs. - Inspecting data stored in Feishu sheets and multidimensional bitables directly within an AI conversation. - Extracting reference notes and slides from Feishu wiki and presentation links for automated summarization.

Key facts

• Open Source
• https://github.com/630156145/weather-server-javascript
• Maps, Weather & Local Data
• forecast, meteorology, nws, weather

Part of MCP Servers

What can Weather do?

Weather provides AI assistants with tools to fetch active US weather alerts by state code and retrieve point weather forecasts using geographic coordinates via the National Weather Service. It also integrates with the Feishu platform to read text content from docs, sheets, slides, bitable bases, and wikis directly inside the agent workflow.

How do I install Weather?

To install Weather, clone the repository from GitHub and install dependencies with npm install. If you plan to use Feishu document features, copy the environment template to create a .env file containing your Feishu App ID and App Secret. You can then run the server via stdio using npm start or through SSE.

Which MCP clients work with Weather?

Weather works with any client supporting the Model Context Protocol, including Claude Desktop and Cursor. It supports standard input and output transport for local desktop configurations as well as HTTP Server-Sent Events on a configurable port for networked or web-based MCP client connections.

Does Weather require an API key for weather data?

No API key is required for weather functionality. The server connects directly to the public US National Weather Service API, which is freely accessible. However, accessing Feishu documents requires creating a custom Feishu application and configuring authentication credentials in your local environment file.

Is Weather open source?

Yes, Weather is open-source software distributed under the ISC License. The full source code is hosted on GitHub, allowing developers to review implementation details, adapt the ES Modules server to custom workflows, and deploy it freely across local or server environments.