WxDecoder is a situational awareness support tool that aggregates real-time aviation data and uses OpenAI to generate plain-English condition reports. It deciphers raw METARs, TAFs, and NOTAMs based on your specific aircraft profile to help enhance your preflight situational awareness.
Note: AI normalizes data and can make errors. Always verify with official sources.
Try it yourself: Live Site
Current Version: v0.75
GitHub: thebronway/WxDecoder
I am a pilot, and honestly, deciphering the raw text of NOTAMs, Airspace, METARs, and TAFs is a massive pain. It is tedious, easy to miss something important, and feels outdated. I wanted a tool that could quickly decode that raw data for me and aid in my normal preflight.
My main goal was to dive into AI engineering and learn how to interact with LLMs. I wanted to see if I could chain together Live APIs (FAA data), LLMs (OpenAI), and Python to solve a real problem I face during preflight. I wanted to build a decoder that reasons through data rather than just displaying it.
- AI Summary: Generates a Briefing Overview covering VFR/IFR status, crosswind components, and runway vectors based on your specific aircraft crosswind tolerance.
- Smart Caching: Intelligent caching strategy that respects METAR update cycles (clears at :50 past the hour) to ensure quality data freshness.
- Smart NOTAMs: Filters through hundreds of raw notices to identify and translate critical hazards (closures, lighting) into plain English.
- Airspace Alerts: Checks your proximity to permanent restricted zones (DC SFRA, P-40, Disney, etc.).
- Vector Math: Automatically calculates crosswind components and compares it to the available runways via onboard logic.
- Contextual Reporting: Integrated feedback tools allow pilots to flag hallucinations, automatically capturing the exact METAR/TAF snapshot for debugging.
- Kiosk Mode: A 16:9 full-screen display optimized for TVs in flight schools and FBOs. It automatically polls for new weather data every minute and refreshes the analysis instantly without user interaction.
- API Costs: The app includes a custom rate limiter (5 requests / 30 minutes) to keep OpenAI costs manageable while in Active Development.
- The rate limit will be normalized to 5 requests / 5 minutes in the future.