iOS app that turns plane spotting into a collection game. Point your phone at a plane in the sky; an AR overlay identifies the aircraft, flight, and airline using live ADS-B data; catch it to your collection.
Status: in TestFlight beta, preparing the v1 App Store launch. The catch flow, local Hangar collection, scoring/rarity economy, trophies, leaderboards, and the production backend are all live. See PLAN.md §9 for the ranked backlog and current status, CHANGELOG.md for history, and CLAUDE.md for working conventions.
- Make plane spotting feel like a game, not a research project
- Reward repeat engagement
- Build a community of avgeeks and casual users alongside each other
- Replacing serious flight trackers (Flightradar24, FlightAware)
- Indoor / non-AR identification
- Multiplayer, trading, Android
iOS app, built via Xcode (⌘R) on a physical iPhone. Simulator can't provide GPS/compass/camera, so device is required.
For tighter iteration there is also a Bash-driven deploy loop: bin/deploy builds, installs over the paired iPhone's wireless link, and launches the app. bin/log-start / bin/log-tail capture the device syslog via idevicesyslog (system-emitted lines about the app; the app's own os_log output still needs Xcode's console). See CLAUDE.md for details.
ADS-B data comes from the Tailspot backend (api.tailspot.app), so the app needs no per-device credentials — just build and run.
Unit tests via xcodebuild test -project ios/Tailspot/Tailspot.xcodeproj -scheme Tailspot -destination 'platform=iOS Simulator,name=iPhone 17,OS=latest' -only-testing:TailspotTests.
See PLAN.md §8. The two deployables are ios/Tailspot/ (the app) and backend/ (the Fly.io API at api.tailspot.app); web/ serves the static legal/marketing pages, tools/ holds dataset generators and eval harnesses, and docs/ carries plans, runbooks, and legal text (historical material in docs/archive/).