Enterprise-grade AI Platform with Multi-Agent Architecture, Voice Integration, and Extensible Tools
Developed by: الحاج ياسر (Hajj Yasser)
Nova is not just another chatbot. It's a production-ready AI platform featuring:
- 🤖 Multi-Agent System - 200+ extensible AI agents
- 🎙️ Voice Engine - Speech-to-Text, Text-to-Speech, Voice Cloning (with consent)
- 🔧 Tool Registry - 200+ extensible tools
- 🧠 Memory System - Conversation, User, Project, Long-term memory
- 🌐 Multi-Provider Support - OpenAI, Anthropic, Google Gemini, Ollama, Local Models
- 📱 Cross-Platform - Web (PWA), iOS, Android, macOS, Windows, Linux
- 🔐 Enterprise Security - Auth, Authorization, Rate Limiting, Sandboxing
- 🏗️ Scalable Architecture - REST API, WebSocket, MCP support
nova-ai/
├── packages/
│ ├── types/ # Shared TypeScript types
│ ├── core/ # Nova Core Engine
│ ├── backend/ # Node.js API Server
│ └── frontend/ # React Web Application
├── docs/ # Documentation
├── docker-compose.yml # Local development setup
├── package.json # Monorepo configuration
└── README.md
- Phase 1: Nova Core + Chat + Provider Abstraction + Basic UI
- Phase 2: Agents + Tools + Memory System
- Phase 3: VoiceStudio Integration (STT, TTS, Voice Conversation)
- Phase 4: Translation + File Processing + OCR
- Phase 5: GitHub Agent + Coding Agent
- Phase 6: App Builder Agent
- Phase 7: Marketplace + Admin Panel + Projects
- Phase 8: PWA + Desktop + Mobile Packaging
# Install dependencies
npm run setup
# Start development server
npm run dev
# Run tests
npm run test
# Build for production
npm run build- React 18+ with TypeScript
- Next.js (optional for SSR)
- Tailwind CSS
- Zustand (State Management)
- React Query (Data Fetching)
- Node.js with Express.js
- TypeScript
- PostgreSQL with Prisma ORM
- Redis (Caching & Queues)
- Docker & Docker Compose
- Provider Abstraction Layer
- OpenAI, Anthropic, Google Gemini APIs
- Ollama for Local Models
- VoiceStudio Adapter
- Web Audio API
- ✅ Nova Core architecture
- ✅ Provider abstraction (OpenAI, Anthropic, Gemini, Ollama)
- ✅ Basic chat interface
- ✅ Conversation memory
- ✅ Error handling & logging
- ✅ API structure
- 🔄 25 core agents
- 🔄 Tool registry system
- 🔄 Memory layer (conversation, user, project)
- 🔄 Agent marketplace UI
- 📅 VoiceStudio integration
- 📅 Speech-to-Text
- 📅 Text-to-Speech
- 📅 Real-time voice conversation mode
- 📅 Translation Agent
- 📅 PDF/Document processing
- 📅 OCR capabilities
- 📅 File upload & analysis
- 📅 GitHub authentication & integration
- 📅 Repository analysis
- 📅 Code review capabilities
- 📅 Pull request generation
- 📅 App Builder Agent
- 📅 Frontend generation
- 📅 Backend scaffolding
- 📅 Database schema creation
- 📅 Agent Marketplace
- 📅 Tool Marketplace
- 📅 Admin Dashboard
- 📅 Project Management
- 📅 PWA configuration
- 📅 Desktop app (Electron)
- 📅 Mobile apps (React Native)
| Component | Status | Notes |
|---|---|---|
| Nova Core | ✅ DONE | Provider abstraction, conversation handling |
| Backend API | ✅ DONE | REST endpoints, database schema |
| Frontend UI | 🔄 IN PROGRESS | Chat interface, basic layout |
| Agents System | 🔄 IN PROGRESS | Registry, execution engine |
| Voice Engine | ⏸️ BLOCKED | Awaiting VoiceStudio API documentation |
| GitHub Integration | ⏸️ BLOCKED | Awaiting OAuth setup |
| Admin Panel | ⏸️ BLOCKED | Awaiting Phase 7 |
Base URL: http://localhost:3001/api
Endpoints:
POST /chat - Send message and get response
GET /agents - List available agents
GET /tools - List available tools
POST /agents/:id/run - Execute specific agent
POST /tools/:id/run - Execute specific tool
GET /conversation/:id - Get conversation history
DELETE /conversation/:id - Clear conversation
POST /memory/:type - Save to memory
GET /memory/:type - Retrieve from memory
Create .env.local:
# Backend
NODE_ENV=development
PORT=3001
DATABASE_URL=postgresql://user:password@localhost:5432/nova
REDIS_URL=redis://localhost:6379
# AI Providers
OPENAI_API_KEY=sk_...
ANTHROPIC_API_KEY=sk_...
GOOGLE_GEMINI_KEY=...
OLLAMA_BASE_URL=http://localhost:11434
# Voice
VOICESTUDIO_API_KEY=...
VOICESTUDIO_BASE_URL=...
# Security
JWT_SECRET=your_secret_key
RATE_LIMIT_REQUESTS=100
RATE_LIMIT_WINDOW=15m
# Frontend
VITE_API_URL=http://localhost:3001
VITE_WS_URL=ws://localhost:3001- ✅ Input validation & sanitization
- ✅ Rate limiting per user/IP
- ✅ SSRF protection
- ✅ Sandboxed tool execution
- ✅ Secrets management (no API keys in frontend)
- ✅ JWT authentication
- ✅ Audit logging for sensitive operations
- ✅ CORS configuration
- API response time: < 200ms (excluding AI processing)
- Chat message processing: < 5s (with streaming)
- Voice recognition: < 2s
- Voice generation: < 3s
- UI load time: < 1.5s
This is currently a solo project by الحاج ياسر. For contributions, please open an issue first to discuss changes.
MIT - See LICENSE file
For issues, feature requests, or questions, please open a GitHub issue.
Made with ❤️ by الحاج ياسر