Weather MCP Server

An MCP server that provides:

Data is sourced from the public NWS API at https://api.weather.gov.

Tools

Tool	Description	Inputs
`get_alerts`	Fetch active NWS alerts for a US state	`state` (string, two-letter code)
`get_forecast`	Fetch short-term forecast (next 5 periods) for coordinates	`latitude` (number), `longitude` (number)

Install dependencies:

pip install .

Run the MCP server (stdio transport):

python -m weather.weather

Your MCP client should be configured to launch the server via the package entrypoint or the above module path.

See server.json for registry metadata including name, version, tools, and entrypoint configuration.

Authenticate with publisher (GitHub namespace):
```
mcp-publisher login github
```
Create and push repo to GitHub (see steps below).
(Optional) Publish to PyPI if distributing as a package.
Publish to MCP registry:
```
mcp-publisher publish
```

Verify:

curl "https://registry.modelcontextprotocol.io/v0/servers?search=io.github.vtiwari/weather-mcp"

If this directory is not yet a git repo:

git init
git add .
git commit -m "Initial commit: Weather MCP server"

Create repo (GitHub CLI) and push:

gh repo create vtiwari/weather-mcp --public --source . --remote origin --push

If not using GitHub CLI, create the repo manually via the GitHub web UI, then:

git remote add origin https://github.com/vtiwari/weather-mcp.git
git branch -M main
git push -u origin main

Tag version for release consistency:

git tag v0.1.0
git push origin v0.1.0

To distribute via PyPI, ensure pyproject.toml includes build backend and metadata (authors, license). Then:

pip install build twine
python -m build
twine upload dist/*

MIT (adjust if different).

This server uses public NOAA/NWS endpoints. Respect API usage guidelines and rate limits.