Athena Weather MCP
Exposes live weather data from Open-Meteo and NWS, providing tools to explore forecasts and list active severe weather alerts. It renders an embedded interactive widget for the Athena platform, with no API keys required.
README
đŠī¸ Weather Alert & Forecast Explorer â Athena MCP Agent
An MCP server that exposes live open weather data and renders it as an embedded interactive widget inside Athena.
Built for the Athena AI Interview Challenge â subject: Weather alert and forecast explorer.
Live endpoints
| URL | |
|---|---|
| MCP endpoint | https://athena-weather-mcp-mocha.vercel.app/mcp |
| Widget preview (browser) | /widget?location=Burlington,Vermont |
| Health | /health |
No authentication â the endpoint is public so any MCP host can connect directly.
Real open/public data sources
No API keys, no auth, no scraping â both are official public feeds.
| Source | Use | Coverage |
|---|---|---|
| Open-Meteo Forecast API | Current conditions, 7-day daily forecast, hourly series | Global |
| Open-Meteo Geocoding API | Free-text location â coordinates | Global |
| US National Weather Service API | Active severe weather alerts/warnings | United States |
Alert coverage is stated honestly in the payload (alertCoverage) rather than
silently returning an empty list outside the US.
MCP tools
explore_weather
Opens the interactive explorer for any location on Earth.
{ "location": "Tampa, Florida", "days": 7 }
Returns a text summary (for the model), structuredContent (for programmatic use),
and the interactive widget.
list_active_alerts
Lists currently active NWS alerts, filterable by state code and severity â used to
find where severe weather is happening, then drill in with explore_weather.
{ "area": "TX", "severity": "Severe" }
The widget â interactions
The widget ships three meaningful interactions (two were required):
- Day selector â click any of the 7 forecast day cards to pivot the entire hourly view to that day. Keyboard accessible (Enter/Space).
- Metric toggle â switch the hourly chart between Temperature, Rain %, and Wind. The chart re-renders as a line or bar series with an axis scaled to the selected metric.
- Expandable alerts â click any active NWS alert to expand the full official description, protective-action instructions, onset/expiry window, urgency and certainty.
Hovering the chart also scrubs a per-hour readout (temp / precip / wind).
Selecting a day or expanding an alert additionally emits a follow-up prompt to the host so the conversation can continue from what the user clicked.
Design notes
Host-convention hedging. MCP hosts disagree on how a tool returns UI. Each
explore_weather result therefore carries the widget in every known convention
at once â an mcp-ui embedded resource in the content array, plus
_meta.ui.resourceUri (Anthropic Apps) and _meta["openai/outputTemplate"]
(OpenAI Apps SDK). Unknown _meta keys are ignored by spec, so this costs
nothing and removes a single point of failure.
Self-contained widget. The full dataset is inlined as JSON at render time and every interaction runs client-side. The widget makes zero network calls and needs zero host cooperation to be fully interactive â important because iframe CSP and host-messaging support vary widely. Host messaging is layered on as enhancement, never as a dependency.
Graceful degradation. Hosts with no widget support still get a complete text
summary plus structuredContent, so the model can answer regardless.
Safety. NWS alert text is third-party input and is rendered exclusively via
textContent / DOM construction â never innerHTML.
Run locally
npm run dev # http://localhost:3000/mcp
Smoke test:
curl -s http://localhost:3000/mcp \
-H 'content-type: application/json' \
-d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"explore_weather","arguments":{"location":"Miami, Florida"}}}'
Open the widget standalone in a browser:
http://localhost:3000/widget?location=Tampa,Florida
Stack
Zero runtime dependencies. Hand-rolled JSON-RPC 2.0 over Streamable HTTP
(POST /mcp, with SSE framing when the client's Accept header asks for it),
deployed as a Node serverless function on Vercel.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.