Skip to main content

Anthropic Claude + RevenueCat + Firebase in a React Native boilerplate

Does AI Mobile Launcher support this combo?

This is a two-graft combo. The boilerplate does not ship Anthropic Claude (only OpenAI + Gemini in AI Pro) and does not ship Firebase as the database (Supabase is the default). Both grafts are evening-scale work, not week-scale, because the patterns are already there: the OpenAI REST client mirrors to Anthropic, and Firebase Analytics + Crashlytics + Remote Config + FCM are already wired so adding Auth + Firestore is the same google-services.json.

The stack

Setup in five steps

1. Clone Standard

git clone <standard-tier-repo>
cd ai-mobile-launcher-standard
pnpm install

2. Anthropic REST client

// src/features/ai/api/anthropic.api.ts
// POST https://api.anthropic.com/v1/messages
// Headers:
//   x-api-key: <ANTHROPIC_API_KEY>
//   anthropic-version: 2023-06-01
//   content-type: application/json
# .env (server proxy)
ANTHROPIC_API_KEY=sk-ant-...
ANTHROPIC_MODEL=claude-haiku-4-5

3. Firebase Auth + Firestore

pnpm add @react-native-firebase/auth @react-native-firebase/firestore
# Replace src/features/auth/services with Firebase versions
# Replace src/store/api/<feature>.api.ts queryFn with Firestore reads

4. RevenueCat env

# .env
EXPO_PUBLIC_REVENUECAT_IOS_KEY=appl_...
EXPO_PUBLIC_REVENUECAT_ANDROID_KEY=goog_...

5. Real-device dev build

eas build --profile development --platform ios

Why this combo works

Claude + Firestore is the right combo for chat-heavy apps with long histories. Firestore listeners handle the real-time UI; Claude's 200K context handles the "remember what the user said three days ago" problem without summarization tricks.

The Firebase Analytics + Crashlytics + Remote Config stack that already ships with the boilerplate is genuinely strong for consumer mobile: funnel analytics, crash-free user rates, remote feature flags. Pairing it with Firestore as the data layer keeps everything in one Google Cloud project, one billing account, one IAM model.

Claude Haiku is the cheapest frontier-class model on quality-per-dollar for chat workloads today. Sonnet is the escalation when reasoning depth matters.

What it costs at scale

Line item1K MAU100K MAU
Claude Haiku (10 chats, 500 tokens)~$5~$500
RevenueCat$0~$200
Firestore~$0~$300-600
Firebase Auth$0~$50
Total (excl. store fees)~$5~$1050-1350

What this combo does NOT cover

Get this combo

Standard tier ($99) is the cleaner base because you are grafting two things. The AI Pro tier ($199) gives you the REST client pattern to mirror.

Standard tier ($99)AI Pro tier ($199)

Related combos