A cognitive architecture platform for persistent organizational knowledge with AI-powered governance, compliance frameworks, and blockchain integration.
Forge V3 consists of three API services and a React frontend:
| Service | Port | Description |
|---|---|---|
| Cascade API | 8001 | Core engine - capsules, governance, overlays, system monitoring |
| Compliance API | 8002 | GDPR/privacy compliance - DSARs, consent, breach notification, AI governance |
| Virtuals API | 8003 | Blockchain integration - agents, tokenization, ACP, revenue tracking |
| Frontend | 5173 | React 19 + TypeScript + Tailwind v4 dashboard |
- Python 3.11+
- Node.js 18+ (for frontend)
- Neo4j Database (cloud or local)
- Redis (optional, for caching)
git clone https://github.com/SunFlash12/ForgeV3.git
cd "Forge V3"cd forge-cascade-v2
python -m venv venv
# Windows
venv\Scripts\activate
# Linux/Mac
source venv/bin/activate
pip install -r requirements.txtCreate a .env file in forge-cascade-v2/:
# Neo4j Database
NEO4J_URI=neo4j+s://your-instance.databases.neo4j.io
NEO4J_USERNAME=neo4j
NEO4J_PASSWORD=your-password
# Security
JWT_SECRET_KEY=your-secure-secret-key-min-32-chars
JWT_ALGORITHM=HS256
ACCESS_TOKEN_EXPIRE_MINUTES=60
# Redis (optional)
REDIS_URL=redis://localhost:6379
# Admin seed password
SEED_ADMIN_PASSWORD=your-secure-admin-password# Set up database schema
python scripts/setup_db.py
# Seed initial data (admin user, sample data)
SEED_ADMIN_PASSWORD="your-password" python scripts/seed_data.pyOption A: Start all servers (recommended)
# Python (cross-platform)
python start_all_servers.py
# Windows
start_servers.bat
# Linux/Mac
./start_servers.shOption B: Start individually
# Terminal 1 - Cascade API
python -m uvicorn forge.api.app:app --host 0.0.0.0 --port 8001
# Terminal 2 - Compliance API
python run_compliance.py
# Terminal 3 - Virtuals API
python run_virtuals.pycd frontend
# Install dependencies
npm install
# Copy environment template
cp .env.example .env
# Start development server
npm run devThe frontend will be available at http://localhost:5173
# 1. Copy environment template
cp .env.example .env
# 2. Edit .env with your configuration
# - Set NEO4J_URI, NEO4J_USERNAME, NEO4J_PASSWORD
# - Set JWT_SECRET_KEY (min 32 chars)
# - Set SEED_ADMIN_PASSWORD
# 3. Build and start all services
docker-compose up -d --build
# 4. Initialize database (first time only)
docker-compose --profile setup run --rm db-setup
docker-compose --profile setup run --rm db-seed
# 5. Access the application
# - Frontend: http://localhost
# - Cascade API: http://localhost:8001
# - Compliance API: http://localhost:8002
# - Virtuals API: http://localhost:8003| Service | Container | Port | Description |
|---|---|---|---|
| cascade-api | forge-cascade-api | 8001 | Core Forge engine |
| compliance-api | forge-compliance-api | 8002 | GDPR compliance |
| virtuals-api | forge-virtuals-api | 8003 | Blockchain integration |
| frontend | forge-frontend | 80 | React dashboard |
| redis | forge-redis | 6379 | Caching layer |
# View logs
docker-compose logs -f
# View specific service logs
docker-compose logs -f cascade-api
# Stop all services
docker-compose down
# Stop and remove volumes
docker-compose down -v
# Rebuild specific service
docker-compose up -d --build cascade-api
# Scale services (if needed)
docker-compose up -d --scale compliance-api=2For production, consider:
- Use external Neo4j: Point
NEO4J_URIto a managed Neo4j instance - Use external Redis: For high availability, use Redis Cluster or managed Redis
- Add reverse proxy: Use Traefik or nginx-proxy for SSL termination
- Set resource limits: Add
deploy.resourcesin docker-compose.yml
Example production override (docker-compose.prod.yml):
version: '3.8'
services:
cascade-api:
deploy:
resources:
limits:
cpus: '2'
memory: 2G
environment:
- LOG_LEVEL=WARNINGcd forge-cascade-v2/frontend
npm run buildProduction files will be in dist/. Deploy to any static hosting.
# With Gunicorn (Linux/Mac)
gunicorn forge.api.app:app -w 4 -k uvicorn.workers.UvicornWorker -b 0.0.0.0:8001
# With Uvicorn
uvicorn forge.api.app:app --host 0.0.0.0 --port 8001 --workers 4Once running, interactive API docs are available at:
- Cascade API: http://localhost:8001/docs
- Compliance API: http://localhost:8002/docs
- Virtuals API: http://localhost:8003/docs
cd "Forge V3"
SEED_ADMIN_PASSWORD="your-password" python test_forge_v3_comprehensive.pycd forge-cascade-v2
SEED_ADMIN_PASSWORD="your-password" python test_ui_integration.pyForge V3/
├── forge-cascade-v2/ # Main application
│ ├── forge/ # Core Python package
│ │ ├── api/ # FastAPI routes
│ │ │ ├── routes/ # API endpoints
│ │ │ ├── app.py # Main FastAPI app
│ │ │ └── middleware.py # Auth, CORS, rate limiting
│ │ ├── database/ # Neo4j client
│ │ ├── immune/ # Circuit breakers, anomaly detection
│ │ ├── kernel/ # Event system, pipeline, overlays
│ │ ├── models/ # Pydantic models
│ │ ├── overlays/ # Governance, ML, security overlays
│ │ ├── repositories/ # Data access layer
│ │ ├── security/ # Auth, tokens, passwords
│ │ └── services/ # Business logic
│ ├── frontend/ # React frontend
│ │ ├── src/
│ │ │ ├── api/ # API client
│ │ │ ├── components/ # React components
│ │ │ ├── pages/ # Page components
│ │ │ ├── stores/ # Zustand stores
│ │ │ └── types/ # TypeScript types
│ │ └── dist/ # Production build
│ ├── scripts/ # Setup and utility scripts
│ ├── run_compliance.py # Standalone compliance server
│ ├── run_virtuals.py # Standalone virtuals server
│ └── start_all_servers.py # Multi-server launcher
├── forge_virtuals_integration/ # Virtuals Protocol SDK
├── Documentation/ # Project documentation
└── test_forge_v3_comprehensive.py # Full test suite
- Knowledge Capsules: Versioned, typed knowledge units with lineage tracking
- 7-Phase Pipeline: Validation → Enrichment → Classification → Security → Governance → Storage → Indexing
- Overlay System: Pluggable processing modules (ML, security, governance)
- Event System: Pub/sub for system-wide event handling
- Ghost Council: AI ethics committee for governance decisions
- DSAR Management: Data Subject Access Requests (GDPR Article 15-22)
- Consent Tracking: Granular consent management with GPC support
- Breach Notification: 72-hour breach reporting workflow
- AI Governance: EU AI Act compliance, human-in-the-loop reviews
- Agent Management: Create and run AI agents
- Tokenization: Bonding curves, governance tokens
- ACP Protocol: Agent Commerce Protocol for service offerings
- Revenue Tracking: DCF valuation, revenue analytics
- JWT Authentication: Access + refresh tokens in httpOnly cookies
- CSRF Protection: Token-based CSRF prevention
- Trust Hierarchy: Role-based access with trust levels
- Rate Limiting: Redis-backed request throttling
- Circuit Breakers: Automatic failure isolation
After running seed_data.py:
| Username | Role | Description |
|---|---|---|
| admin | admin | System administrator |
| oracle | user | High-trust user |
| developer | user | Standard user |
| analyst | user | Standard user |
Passwords are set via SEED_ADMIN_PASSWORD environment variable.
- Verify
NEO4J_URI,NEO4J_USERNAME,NEO4J_PASSWORDin.env - Check Neo4j instance is running and accessible
- For cloud instances, ensure IP is whitelisted
# Find and kill process on port (Windows)
netstat -ano | findstr :8001
taskkill /PID <pid> /F
# Linux/Mac
lsof -i :8001
kill -9 <pid>- Ensure
.envhas correctVITE_API_URL=http://localhost:8001/api/v1 - Restart frontend dev server after changing
.env - Check CORS settings in backend
Proprietary - All rights reserved.
For issues and feature requests, visit: https://github.com/SunFlash12/ForgeV3/issues