alexei.dolgolyov/personal-ai-assistant
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
8 Commits 1 Branch 0 Tags
04e3ae831927a02f194d151b9f6fa7fa3d1907f4
Commit Graph

8 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
dolgolyov.alexei
04e3ae8319 Fix Phase 5 review issues: mark_all_read, datetime, WS payload, logging 
Review fixes:
- mark_all_read now filters status IN (sent, delivered), preventing
  pending notifications from being incorrectly marked as read
- schedule_notification tool: force UTC on naive datetimes, reject
  past dates (send immediately instead)
- WebSocket notification payload now includes all fields matching
  NotificationResponse (user_id, channel, status, scheduled_at, etc.)
- Centralized _serialize_notification helper in notification_sender
- notification_sender: per-notification error handling with rollback
- health_review: moved import json to module top level

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
 2026-03-19 14:14:08 +03:00
dolgolyov.alexei
ada7e82961 Phase 5: Notifications — WebSocket, APScheduler, AI tool, health review 
Backend:
- Notification model + Alembic migration
- Notification service: CRUD, mark read, unread count, pending scheduled
- WebSocket manager singleton for real-time push
- WebSocket endpoint /ws/notifications with JWT auth via query param
- APScheduler integration: periodic notification sender (every 60s),
  daily proactive health review job (8 AM)
- AI tool: schedule_notification (immediate or scheduled)
- Health review worker: analyzes user memory via Claude, creates
  ai_generated notifications with WebSocket push

Frontend:
- Notification API client + Zustand store
- WebSocket hook with auto-reconnect (exponential backoff)
- Notification bell in header with unread count badge + dropdown
- Notifications page with type badges, mark read, mark all read
- WebSocket initialized in AppLayout for app-wide real-time updates
- Enabled notifications nav in sidebar
- English + Russian translations

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
 2026-03-19 13:57:25 +03:00
dolgolyov.alexei
8b8fe916f0 Phase 4: Documents & Memory — upload, FTS, AI tools, context injection 
Backend:
- Document + MemoryEntry models with Alembic migration (GIN FTS index)
- File upload endpoint with path traversal protection (sanitized filenames)
- Background document text extraction (PyMuPDF)
- Full-text search on extracted_text via PostgreSQL tsvector/tsquery
- Memory CRUD with enum-validated categories/importance, field allow-list
- AI tools: save_memory, search_documents, get_memory (Claude function calling)
- Tool execution loop in stream_ai_response (multi-turn tool use)
- Context assembly: injects critical memory + relevant doc excerpts
- File storage abstraction (local filesystem, S3-swappable)
- Secure file deletion (DB flush before disk delete)

Frontend:
- Document upload dialog (drag-and-drop + file picker)
- Document list with status badges, search, download (via authenticated blob)
- Document viewer with extracted text preview
- Memory list grouped by category with importance color coding
- Memory editor with category/importance dropdowns
- Documents + Memory pages with full CRUD
- Enabled sidebar navigation for both sections

Review fixes applied:
- Sanitized upload filenames (path traversal prevention)
- Download via axios blob (not bare <a href>, preserves auth)
- Route ordering: /search before /{id}/reindex
- Memory update allows is_active=False + field allow-list
- MemoryEditor form resets on mode switch
- Literal enum validation on category/importance schemas
- DB flush before file deletion for data integrity

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
 2026-03-19 13:46:59 +03:00
dolgolyov.alexei
03afb7a075 Phase 3: Skills & Context — skill system, personal context, context layering 
Backend:
- Skill model + migration (with FK on chats.skill_id)
- Personal + general skill CRUD services with access isolation
- Admin skill CRUD endpoints (POST/GET/PATCH/DELETE /admin/skills)
- User skill CRUD endpoints (POST/GET/PATCH/DELETE /skills/)
- Personal context GET/PUT at /users/me/context
- Extended context assembly: primary + personal context + skill prompt
- Chat creation/update now accepts skill_id with validation

Frontend:
- Skill selector dropdown in chat header (grouped: general + personal)
- Reusable skill editor form component
- Admin skills management page (/admin/skills)
- Personal skills page (/skills)
- Personal context editor page (/profile/context)
- Updated sidebar: Skills, My Context nav items + admin skills link
- English + Russian translations for all skill/context UI

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
 2026-03-19 12:55:02 +03:00
dolgolyov.alexei
70469beef8 Phase 2: Chat & AI Core — Claude API streaming, chat UI, admin context 
Backend:
- Chat, Message, ContextFile models + Alembic migration
- Chat CRUD with per-user limit enforcement (max_chats)
- SSE streaming endpoint: saves user message, streams Claude response,
  saves assistant message with token usage metadata
- Context assembly: primary context file + conversation history
- Admin context CRUD (GET/PUT with version tracking)
- Anthropic SDK integration with async streaming
- Chat ownership isolation (users can't access each other's chats)

Frontend:
- Chat page with sidebar chat list + main chat window
- Real-time SSE streaming via fetch + ReadableStream
- Message bubbles with Markdown rendering (react-markdown)
- Auto-growing message input (Enter to send, Shift+Enter newline)
- Zustand chat store for streaming state management
- Admin primary context editor with unsaved changes warning
- Updated routing: /chat, /chat/:chatId, /admin/context
- Enabled Chat and Admin sidebar navigation
- English + Russian translations for all new UI

Infrastructure:
- nginx: disabled proxy buffering for SSE support
- Added ANTHROPIC_API_KEY and CLAUDE_MODEL to config

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
 2026-03-19 12:38:30 +03:00
dolgolyov.alexei
7c752cae6b Phase 1: Foundation — backend auth, frontend shell, Docker setup 
Backend (FastAPI):
- App factory with async SQLAlchemy 2.0 + PostgreSQL
- Alembic migration for users and sessions tables
- JWT auth (access + refresh tokens, bcrypt passwords)
- Auth endpoints: register, login, refresh, logout, me
- Admin seed script, role-based access deps

Frontend (React + TypeScript):
- Vite + Tailwind CSS + shadcn/ui theme (health-oriented palette)
- i18n with English and Russian translations
- Zustand auth/UI stores with localStorage persistence
- Axios client with automatic token refresh on 401
- Login/register pages, protected routing
- App layout: collapsible sidebar, header with theme/language toggles
- Dashboard with placeholder stats

Infrastructure:
- Docker Compose (postgres, backend, frontend, nginx)
- Nginx reverse proxy with WebSocket support
- Dev override with hot reload

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
 2026-03-19 12:25:02 +03:00
dolgolyov.alexei
5bdc296172 Add README and ignore .claude directory 
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
 2026-03-19 11:32:52 +03:00
dolgolyov.alexei
96a1aabafc Initial commit: project plan and gitignore 
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
 2026-03-19 11:30:58 +03:00