OMNI-AI is a cutting-edge multi-agent autonomous AI system featuring neuro-symbolic architecture, advanced security, and specialized intelligent agents for complex task execution.
- NEXUS Orchestrator: Central coordination system for task decomposition and agent management
- 9 Specialized Agents:
- VERITAS: Truth verification and fact-checking
- LEX-Core: Legal and compliance analysis
- CERBERUS: Security monitoring and threat detection
- FORGE: Engineering and design
- VITA: Biological and medical analysis
- MUSE: Creative content generation
- ARES: Strategic planning and optimization
- LUDUS: Simulation and gaming
- ARGUS: Monitoring and analytics
- AEGIS Guardian Layer: Real-time input/output filtering and content censorship
- Omni-Auth: Multi-level authentication system (Level 1-3)
- Post-Quantum Cryptography: Kyber (FIPS 203) KEM and Dilithium (FIPS 204) signatures
- Threat Detection: Automated security event monitoring and response
- Working Memory: Fast, temporary storage with LRU cache and TTL support
- Long-Term Memory: Persistent, encrypted knowledge graph using Neo4j
- Vector Store: Semantic similarity search using Pinecone
- CAD & Blueprint Generator: Parametric 3D modeling and design automation
- Multi-Physics Simulation Engine: OpenFOAM integration for CFD simulations
- Digital Twin Simulator: Real-time simulation of biological and physical systems
- Code Sandbox Environment: Docker-based isolated execution environment
- Combines deep learning (DeepSeek-V3 MoE) with symbolic reasoning
- Enhanced logical consistency and explainability
- Multi-Token Prediction (MTP) for improved reasoning
- Python 3.11 or higher
- Docker and Docker Compose
- Redis
- Neo4j
- (Optional) Pinecone for vector search
- Clone the repository:
git clone https://github.com/vantisCorp/V-AGI.git
cd V-AGI- Create virtual environment:
python -m venv venv
source venv/bin/activate # On Windows: venv\Scripts\activate- Install dependencies:
pip install -r requirements.txt- Configure environment:
cp .env.example .env
# Edit .env with your configuration- Start services with Docker Compose:
docker-compose up -d- Run the application:
python src/main.pyV-AGI/
├── .github/workflows/ # CI/CD pipelines
├── src/
│ ├── agents/ # 9 Specialized agents + base
│ ├── api/ # REST API, WebSocket, messaging
│ ├── memory/ # Working, long-term, vector store
│ ├── nexus/ # NEXUS orchestrator
│ ├── security/ # AEGIS Guardian layer
│ ├── tools/ # CAD, physics, digital twin, sandbox
│ ├── config.py # Application configuration
│ └── main.py # Entry point
├── tests/ # Test suite (840+ tests, 90% coverage)
├── docs/ # Documentation
├── examples/ # Usage examples
├── scripts/ # Setup and utility scripts
├── Dockerfile # Container configuration
└── docker-compose.yml # Multi-service orchestration
Key configuration options in .env:
# Application
APP_NAME=OMNI-AI
ENVIRONMENT=development
# Security
SECRET_KEY=your-secret-key-here
JWT_SECRET_KEY=your-jwt-secret-key-here
# Database
REDIS_URL=redis://localhost:6379/0
NEO4J_URI=bolt://localhost:7687
# AI Models
ANTHROPIC_API_KEY=your-anthropic-api-key
DEEPSEEK_API_KEY=your-deepseek-api-key┌─────────────────────────────────────────────────────────┐
│ User Interface │
│ (Generative UI + AR/VR) │
└────────────────────────┬────────────────────────────────┘
│
┌────────────────────────▼────────────────────────────────┐
│ AEGIS Guardian Layer │
│ (Input/Output Filtering & Censorship) │
└────────────────────────┬────────────────────────────────┘
│
┌────────────────────────▼────────────────────────────────┐
│ NEXUS Orchestrator │
│ (Task Decomposition & Agent Coordination) │
└────────────┬─────────────────────────────┬──────────────┘
│ │
┌────────────▼────────────┐ ┌────────────▼────────────┐
│ Specialized Agents │ │ Memory Systems │
│ (VERITAS, LEX-Core, │ │ (Working, Long-Term, │
│ CERBERUS, FORGE, │ │ Vector Store) │
│ VITA, MUSE, ARES, │ └────────────────────────┘
│ LUDUS, ARGUS) │
└─────────────────────────┘
│
┌────────────▼────────────────────────────────────────────┐
│ Neuro-Symbolic Foundation Layer │
│ (DeepSeek-V3 MoE + Symbolic Reasoning + Knowledge) │
└─────────────────────────────────────────────────────────┘
- Level 1 (Safe Mode/Guest): Basic operations with content filtering
- Level 2 (Specialist): Advanced features with 2FA + biometrics
- Level 3 (Root Mode): Full system access with Golden Key Protocol
- Kyber-1024: Quantum-resistant key encapsulation
- Dilithium-5: Quantum-resistant digital signatures
- AES-256-GCM: Hybrid encryption for data at rest
Run the test suite:
pytest tests/ -vRun with coverage:
pytest tests/ --cov=src --cov-report=html- Agents Implementation
- Specialized Agents Summary
- Security Implementation
- Simulation Tools
- Advanced Tools Summary
- API Integration Summary
This project uses GitHub Actions for continuous integration and deployment. The CI pipeline includes:
- Testing: Automated test execution with coverage reporting
- Linting: Code style checks with Black, isort, and Flake8
- Security: Dependency vulnerability scanning with Safety
- Build: Package building and artifact publishing
# Run tests
pytest tests/ --cov=src
# Run linting
black --check src/ tests/
isort --check-only src/ tests/
flake8 src/ tests/
# Run security check
pip install safety && safety check -r requirements.txtFor private repositories, GitHub Actions requires a paid plan or self-hosted runners. See the Self-Hosted Runner Setup Guide for instructions.
Contributions are welcome! Please follow these guidelines:
- Fork the repository
- Create a feature branch (
git checkout -b feature/amazing-feature) - Commit your changes (
git commit -m 'Add amazing feature') - Push to the branch (
git push origin feature/amazing-feature) - Open a Pull Request
OMNI-AI is designed to comply with:
- EU AI Act: Comprehensive AI regulation framework
- ISO/IEC 42001: AI management system standard
- NIST AI RMF: AI Risk Management Framework
- FIPS 140-3 Level 4: Cryptographic module validation
- ISO/IEC 27001: Information security management
- SOC 2 Type II: Security and availability controls
- HIPAA/RODO: Healthcare and data privacy regulations
- ISO 13485: Medical device quality management
- FedRAMP/DoD IL6: Cloud security authorization
- Python 3.11+, asyncio, pydantic
- FastAPI for REST APIs
- DeepSeek-V3 (Mixture-of-Experts)
- Anthropic SDK (MCP protocol)
- Transformers, PyTorch
- cryptography, argon2-cffi, pyotp
- Post-quantum crypto (Kyber, Dilithium)
- Neo4j (Knowledge Graph)
- Pinecone (Vector Store)
- Redis (Cache & Working Memory)
- OpenFOAM (CFD)
- FreeCAD (CAD operations)
- NumPy, SciPy (Numerical computing)
This project is licensed under the MIT License - see the LICENSE file for details.
OMNI-AI Development Team
- DeepSeek for MoE architecture insights
- Anthropic for tool system and MCP protocol
- xAI for distributed processing patterns
- Google Gemini for advanced AI techniques
For support and questions:
- Open an issue on GitHub
- Contact the development team
- Check the documentation
Note: OMNI-AI is currently in active development. Features and APIs may change as the project evolves.