Claude
bf5c4e5f00
Takes V1 from code-complete toward the closed beta: - Tests: the suite grows from 5 to 29. handler.test.ts covers route dispatch, guest device-id validation and user creation, JWT provider derivation, the paywall flag pass-through, free/premium mode selection, the assistant-final transcript rule (post-unlock delivery), the 402 abuse cap and its premium exemption, and purchase-verify validation. chat.test.ts covers structured-output parsing, premium masking of analysis_ready, the exact Responses API payload (knowledge base + strict JSON schema + the no-manufactured-suspense rule) and error handling. subscription.test.ts covers the Apple adapter (active/expired/wrong-product/sandbox retry on 21007) and the Google adapter (real RS256 JWT signing against a generated key, token exchange, expiry, 410-gone), with fetch and secrets mocked. - EAS: apps/mobile/eas.json with development/preview/production profiles for TestFlight and Play internal-testing builds. - LAUNCH.md: step-by-step path to 20-50 beta users — AWS deploy and secrets, migration, sign-in providers, store products at $5.99/$49.99, EAS builds, CI role, and the beta measurement plan (retention, dialogue depth, upgrade timing, tester interviews). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_0118DaxZR36RpnY524vRqx3z
3.6 KiB
3.6 KiB
Launch checklist — closed beta
The path from this repo to 20–50 test users. Everything below is configuration and store work; the code is done (see Definition of Done in the README).
1. AWS
cd infra && npx cdk bootstrap(once per account/region).npx cdk deploy— note the outputs:ApiUrl,UserPoolId,UserPoolClientId,CognitoDomain.- Fill the
semantika/appsecret in Secrets Manager:OPENAI_API_KEY(required for chat),APPLE_SHARED_SECRETandGOOGLE_SERVICE_ACCOUNT_JSON(required before purchases work). - Run
db/migrations/001_init.sqlagainst the Aurora cluster (credentials in the RDS-managed secret; connect via a bastion or the RDS query editor). - Smoke test:
curl <ApiUrl>/suggestionsreturns the starter list.
2. Sign-in
- Email sign-in works out of the box (Cognito hosted UI).
- Apple: create a Services ID + key in the Apple Developer portal, then
redeploy with CDK context
appleTeamId,appleKeyId,applePrivateKeySecretName. - Google: create an OAuth client in Google Cloud, then redeploy with
googleClientId,googleClientSecret. - For the beta, email-only is acceptable — Apple/Google can land in a later build. Note: Apple's review requires Sign in with Apple if other third-party logins are offered, so enable it before public App Store release.
3. App configuration
- Put the CDK outputs into
apps/mobile/app.json→extra(apiUrl,cognitoDomain,cognitoClientId). - Keep product ids as
semantika_monthly/semantika_yearly.
4. Stores
- App Store Connect: create the app (bundle id
com.semantika.app), add auto-renewable subscriptionssemantika_monthly($5.99/month) andsemantika_yearly($49.99/year) in one subscription group. Generate the App-Specific Shared Secret →APPLE_SHARED_SECRET. - Play Console: create the app (package
com.semantika.app), add the two subscriptions with the same ids and prices. Create a service account with Play Developer API access → its JSON key becomesGOOGLE_SERVICE_ACCOUNT_JSON. - Store copy follows the honest-claims rule: inspiration, not proven effects. Positioning: reflection partner — not therapy.
5. Builds
cd apps/mobile && npx eas build --profile preview --platform all(in-app purchases require a real build, not Expo Go).- iOS: distribute via TestFlight (internal, then external testers).
- Android: distribute via Play Console internal testing track.
- Purchases in test: use TestFlight sandbox accounts / Play license testers — no real charges.
6. CI/CD
- Create the GitHub OIDC deploy role in AWS; set repo secret
AWS_DEPLOY_ROLE_ARN(and optionally theAWS_REGIONvariable). - Merge this branch to
main— CI runs lint/typecheck/tests,deploy.ymldeploys the stack automatically.
7. The beta itself (20–50 users)
Measure before building anything new:
- Do they come back? — active users over time (
usageactivity). - How deep do dialogues go? —
messages_useddistribution. - When do they upgrade? —
subscriptiontransitions relative tocreated_at. - Which starter questions create value? — conversations are never stored, so ask the testers directly (short interviews or a 3-question survey beats analytics here).
Exit criteria for V1 → V2 decisions: a clear picture of retention, dialogue depth, upgrade timing, and which conversation types resonate. Only then open the Version 2 list.