Pilot-developer ships free Aviation MCP server with live METAR data
A commercial pilot and software developer built a free Aviation MCP server that gives AI agents like Claude and Cursor access to live METAR weather, airport info, aircraft specs, and FAA practice questions — no API key required.
Score breakdown
The server fills a concrete gap in the MCP ecosystem by giving any MCP-compatible AI client real-time aviation weather and reference data that LLMs previously could not access without hallucinating or deferring to external sites.
- 01Built by Perufitlife, a commercial pilot and software developer, to fix AI hallucination of live aviation weather data.
- 02Exposes six tools: `get_metar`, `get_airport`, `get_aircraft`, `get_glossary_term`, `practice_questions`, and `quiz_of_the_day`.
- 03`get_metar` returns decoded METAR data including flight category, wind, visibility, temp, dewpoint, and optional TAF for any ICAO airport.
Perufitlife, a commercial pilot and software developer, built an Aviation MCP server to address a concrete gap: AI assistants given aviation weather queries either hallucinate METAR data or redirect users to external websites, because LLMs have no access to live aviation data. The server wraps the author's own free aviation API, Rotate Pilot, as a thin MCP layer, exposing six typed tools: `get_metar` (current decoded METAR including flight category, wind, visibility, temperature, and dewpoint, with optional TAF), `get_airport` (ICAO lookup returning name, IATA code, city, coordinates, elevation, and runways), `get_aircraft` (specs by slug including engines, range, cruise speed, ceiling, MTOW, and type rating), `get_glossary_term`, `practice_questions` (FAA-style exam questions with answers), and `quiz_of_the_day`.
The server runs as an Apify Standby Actor supporting both Streamable HTTP `/mcp` and legacy SSE transports, built with the official `@modelcontextprotocol/sdk`.
The server runs as an Apify Standby Actor supporting both Streamable HTTP `/mcp` and legacy SSE transports, built with the official `@modelcontextprotocol/sdk`. Each tool definition is approximately 10 lines — a name, a JSON schema, and a `buildPath(args)` function that returns the API path — making the implementation deliberately minimal. Configuration requires only adding the Apify Actor URL with a token to an MCP client's server config. The project is published on GitHub under the MIT license and listed on the Apify Store, with the underlying Rotate Pilot API and its OpenAPI spec available at rotatepilot.com/developers. The post frames MCP as the default tool-delivery mechanism for LLMs in 2026 and argues that wrapping any existing API as an MCP server is an underutilized distribution channel for API owners.
Key facts
- 01Built by Perufitlife, a commercial pilot and software developer, to fix AI hallucination of live aviation weather data.
- 02Exposes six tools: `get_metar`, `get_airport`, `get_aircraft`, `get_glossary_term`, `practice_questions`, and `quiz_of_the_day`.
- 03`get_metar` returns decoded METAR data including flight category, wind, visibility, temp, dewpoint, and optional TAF for any ICAO airport.
- 04No API key or signup required; wraps the author's own free Rotate Pilot aviation API.
- 05Runs as an Apify Standby Actor using the official `@modelcontextprotocol/sdk`, supporting Streamable HTTP `/mcp` and legacy SSE.
- 06Each tool definition is approximately 10 lines: a name, a JSON schema, and a `buildPath(args)` function.
- 07Released under the MIT license on GitHub at github.com/Perufitlife/aviation-mcp.
Topics
Summary and scoring are generated automatically from the original article. We always link back to the publisher and never republish images or paywalled content. Last processed Jun 14, 2026 · 09:08 UTC. How this works →