​
Prerequisites

Before installing NeuralTrust with Docker, ensure you have:
  • Docker Engine 20.10+
  • Docker Compose v2.0+
  • 4GB RAM minimum
  • 10GB disk space
  • OpenAI API key (for AI model access)
  • HuggingFace token (provided by NeuralTrust)

​
Quick Start

​
1. Clone the Repository

# Clone the repository
git clone https://github.com/NeuralTrust/TrustGate.git
cd TrustGate

# Clone environment file
cp .env.example .env

​
2. Configure Environment

Edit the .env file with your configuration: 
# Core settings
ENVIRONMENT=production
DOMAIN=your-domain.com
LOG_LEVEL=info

# API Keys
OPENAI_API_KEY=your-openai-key
HUGGINGFACE_TOKEN=your-hf-token

# Security settings
ENABLE_RATE_LIMITING=true
ENABLE_AUTH=true

​
3. Start the Services

# For production deployment
docker compose -f docker-compose.prod.yaml up -d

# For development
docker compose up -d

​
Architecture Components

The deployment consists of two main APIs:
  1. Admin API (Port 8080)
    • Gateway management
    • Configuration management
    • API key management
    • Plugin configuration
  2. Proxy API (Port 8081)
    • Request routing
    • Load balancing
    • Plugin execution

​
Monitoring

Enable monitoring in your configuration: 
# config.yaml
metrics:
  enabled: true                # Enable metrics collection
  enable_latency: true        # Basic latency metrics
  enable_upstream: true       # Upstream latency tracking
  enable_connections: true    # Connection tracking
  enable_per_route: true     # Per-route metrics

​
Available Metrics

MetricDescription
trustgate_requests_totalRequest counts by gateway, method, and status
trustgate_latency_msOverall request processing time
trustgate_detailed_latency_msGranular latency by service/route
trustgate_upstream_latency_msUpstream service latency
trustgate_connectionsActive connection tracking

​
Plugin System

NeuralTrust includes several built-in plugins:

​
Rate Limiting

  • Basic rate limiter
  • Token rate limiter

​
Prompt Moderation

  • Data masking (entities, keywords, regex)
  • Prompt moderation (keywords, regex)

​
Toxicity Detection

  • OpenAI Moderation API
  • Azure Content Safety API

​
Other Plugins

  • External API calls
  • Custom integrations

​
Managing the Deployment

Common management commands: 
# View logs
docker compose logs -f [service]

# Restart services
docker compose restart [service]

# Stop all services
docker compose down

# Update to latest version
docker compose pull
docker compose up -d

​
Troubleshooting

Common issues and solutions:
  1. Service Status
docker compose ps
  1. View Logs
docker compose logs -f [service]
  1. Check Configuration
docker compose config
  1. Resource Usage
docker stats
  1. Network Issues
# Test internal connectivity
docker compose exec api ping proxy

​
Security Features

The deployment includes:
  • Authentication and authorization
  • Rate limiting and token management
  • CORS protection
  • SQL injection prevention
  • Cross-site injection protection
  • Prompt moderation and jailbreak protection

​
Additional Resources

For more information: For support, contact support@neuraltrust.ai