- Admin-configurable AI prompt/quick questions from system_configs DB
- GET /api/v1/ai/quick-questions endpoint for fetching quick questions
- Local FAQ matching for instant responses (avoid AI calls for common Qs)
- AI action extraction: "add customer" intent detected, structured data returned
- Frontend action confirmation card with editable fields, calls customer API on confirm
- NVIDIA provider (stepfun-ai/step-3.5-flash) for faster chat vs deepseek-v4-flash
- Fixed httpx client timeout preventing backend hangs
- Added log_usage calls for auth events (register/login/guest/wechat)
- Admin tabs (users/stats/logs/config) fully functional with real backend
- AiAssistant component added to all tabbar pages
- Switch back to custom tabbar (custom: true) with emoji fix (line-height, font-family)
- Redesign home quick-actions: product/followup/analytics/notifications replace redundant nav items
- Refactor goToPage to auto-detect tabbar vs non-tabbar pages
- Update PROGRESS.md
- Backend: guest UUID format fix, /auth/me guest branch, UUID validation in deps.py, CORS config fix
- Frontend: switch to native tabbar (custom: false), cleanup App.vue, redesign quick-actions with colored icons, conditional wechat login, proxy API requests via Vite