- Added a new onboarding wizard that runs on the first app launch post-login.
- Collects user personalization data (job role, industry, primary use case, tone preference, language) and stores it in encrypted core memory.
- Auto-detects and saves formatting preferences (timezone, time format, date format) in local electron-store.
- Normalizes user free-text inputs via a backend LLM call before persisting.
- Introduced new backend routes for memory updates and normalization.
- Updated frontend components to support the onboarding flow with a chat-bubble aesthetic.
- Added settings section for profile editing and re-running the onboarding process.
- Ensured that the onboarding process is skippable and editable in the settings.
- Implemented verification steps to ensure proper functionality and data handling.
chore: update submodules for waitlist and website
Adds api/app/auth/oauth_providers.py with GoogleOAuthProvider (httpx-based,
no authlib needed) and generate_pkce_pair(). New routes:
GET /auth/oauth/{provider}/authorize and POST /auth/oauth/{provider}/callback
with state/PKCE validation and three-way user resolution (existing OAuth link,
email auto-link, new social-only user). Updates settings.py with
GOOGLE_AUTH_CLIENT_ID/SECRET and OAUTH_REDIRECT_URI.
Also includes Step 1 backend changes (already marked complete in plan):
oauth_accounts table migration, nullable password_hash, avatar_url on User.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
