AI Mobile Launcher
I spent 6 months on the architecture so you don't have to spend 6 weeks on it. A React Native and Expo boilerplate built to work well with Claude Code. The whole stack is included, and AI features are ready.
The problem
Every React Native project starts the same way. You spend the first four to six weeks wiring up auth, state, navigation, payments, analytics, and CI/CD before you write a line of real business logic. And if you're building with Claude Code or Cursor, there's a second problem: a blank project gives the AI nothing to reason from. So it guesses, and on mobile the guesses are usually wrong.
What I built
AI Mobile Launcher is a React Native and Expo boilerplate built to be read by people and AI tools alike. Every folder, rule, and generator is set up so Cursor and Claude Code write correct mobile code on the first prompt, instead of guessing at your patterns.
U-AMOS 2.0 Claude Code skeleton
A 9-file memory bank, rule packs, and step-by-step generators. Cursor and Claude write code that fits the architecture from the start. No guessing, no regressions.
The whole stack, already wired
RevenueCat paywall, Supabase auth, Firebase Analytics, Crashlytics, FCM push, and offline-first sync with conflict resolution. Everything a real app needs, set up and tested.
AI Pro: on-device and cloud LLM
llama.rn runs GGUF models fully offline. Gemini and OpenAI clients handle cloud inference. The user picks the provider in Settings, and six AI screens are ready to ship.
Feature-first architecture
A folder layout AI tools can extend without breaking it. Each feature stands on its own. New screens don't touch the old ones. Built for a codebase that keeps growing.
Three tiers, one codebase
MIT licensed, GitHub
- Expo SDK 55 + RN New Architecture
- Feature-first folder architecture
- Auth + onboarding flows
- Restyle UI + Redux Toolkit
- Basic Cursor rules
Pay once
- Everything in Lite
- Offline-first sync (MMKV)
- RevenueCat paywall + trial
- Supabase auth + RLS
- Firebase + Sentry + U-AMOS 2.0
Pay once
- Everything in Standard
- On-device LLM (llama.rn)
- Gemini + OpenAI cloud LLMs
- Multimodal analyser
- Wire RN Dynamic Onboarding
Technology stack
Start building today
Grab the boilerplate, or let me build your AI-first mobile app from scratch on the same stack, shaped around your product.