MCP server giving AI agents (ChatGPT, Claude, Cursor) live aviation data: decoded METAR weather, airport info, aircraft specs, glossary, FAA practice questions, and a daily pilot quiz. Powered by the free Rotate Pilot API.
Aviation-mcp demonstrates solid tool design with clear naming conventions (all tools start with action verbs: get_, practice_), complete input schemas with proper JSON Schema structure, and descriptive text for all tools and parameters. All 6 tools have descriptions (avg 156 chars, within 10-1024 baseline). Parameters are well-typed with descriptions. However, output schemas are not documented, LLMs cannot predict response structure for downstream chaining. Error handling is basic (generic error messages without recovery guidance). No tool annotations (readOnlyHint, destructiveHint) despite all tools being read-only. Parameter descriptions could be more prescriptive about format/constraints (e.g., 'ICAO code must be exactly 4 uppercase letters').
Get aircraft specifications by slug: engines, seats, range, cruise speed, ceiling, MTOW, type-rating requirement, and description. Example slugs: cessna-172-skyhawk, piper-pa-28-warrior, airbus-a320neo, boeing-737-800.
Get airport information by ICAO code: name, IATA code, city, country, region, coordinates, elevation, and runway count.
Look up the definition of an aviation term from the Rotate Pilot glossary by slug (e.g. squawk, angle-of-attack, v-speeds). Returns the definition, category, and related terms.
Get the current decoded METAR (aviation weather) for an airport by ICAO code. Returns raw report, flight category (VFR/MVFR/IFR/LIFR), wind, visibility, temperature, dewpoint, and more. Optionally include the TAF forecast.
Get FAA-style aviation practice exam questions with options and the correct answer. Optionally filter by subject (e.g. meteorology, regulations, navigation). Great for building study tools or quizzing a user.
Output schemas not documented. LLMs cannot predict response structure (fields, types, nesting) for downstream tool chaining or data extraction. Agents must infer structure from trial calls.
Parameter descriptions lack prescriptive format constraints. 'icao' should state 'exactly 4 uppercase letters (e.g., KJFK)'; 'slug' should document valid format. Current descriptions are permissive and invite invalid input.
Error handling is generic. fetchApi() returns 'Error: <message>' without recovery guidance. LLMs cannot distinguish retryable errors (network timeout) from user-fixable errors (invalid ICAO code) from fatal errors (API down).
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | C | 66 | <=2025-11-25 | v2 |
Get the aviation question of the day — a single FAA-style multiple-choice question with the correct answer and explanation.
No tool annotations. All tools are read-only (safe to retry), but readOnlyHint is not set. Agents cannot infer safety properties and may apply unnecessary caution or retry logic.
practice_questions 'subject' parameter is optional but no enum constraint. Description lists examples ('meteorology, regulations, navigation') but LLMs may hallucinate other values. Should declare enum or validate server-side with clear error.