Configuration

Environment Variables

Variable	Required	Default	Description
`LLM_PROVIDER`	No	`openai`	LLM provider to use (`openai` or `bedrock`)
`OPENAI_API_KEY`	Yes*	—	OpenAI API key (*required when `LLM_PROVIDER=openai`)
`PORT`	No	`3001`	Server port
`MONGODB_URI`	Yes	—	MongoDB connection string
`AUTH_SECRET`	Yes	—	Secret for JWT auth verification
`LOG_LEVEL`	No	`info`	Logging level (`debug`, `info`, `warn`, `error`)

Provider Configuration

OpenAI

LLM_PROVIDER=openai
OPENAI_API_KEY=sk-...

Supported models: gpt-4o, gpt-4o-mini, gpt-4-turbo, gpt-3.5-turbo

AWS Bedrock (Planned)

LLM_PROVIDER=bedrock
AWS_REGION=us-east-1
AWS_ACCESS_KEY_ID=AKIA...
AWS_SECRET_ACCESS_KEY=...
BEDROCK_MODEL_ID=anthropic.claude-3-haiku-20240307-v1:0

Rate Limiting

Rate limits are enforced per-user based on token consumption. Configure via:

RATE_LIMIT_RPM=60          # Requests per minute per user
RATE_LIMIT_TPM=100000      # Tokens per minute per user

Docker Compose

version: '3.8'
services:
  norman-engine:
    build: .
    ports:
      - "3001:3001"
    environment:
      - LLM_PROVIDER=openai
      - OPENAI_API_KEY=${OPENAI_API_KEY}
      - MONGODB_URI=mongodb://mongo:27017/norman-engine
      - AUTH_SECRET=${AUTH_SECRET}
    depends_on:
      - mongo
 
  mongo:
    image: mongo:7
    ports:
      - "27017:27017"
    volumes:
      - mongo_data:/data/db
 
volumes:
  mongo_data:

Getting Started API Reference