gitforai/docker-compose.yml at master · git-for-ai/gitforai · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
# Docker Compose configuration for GitForAI
# Production/CLI usage

version: '3.8'

services:
  gitforai:
    build:
      context: .
      target: runtime
    image: gitforai:latest
    container_name: gitforai-cli

    # Mount local directories
    volumes:
      # Mount repository to analyze (customize path as needed)
      - ./repos:/repos:ro  # Read-only access to repositories
      - ./output:/output   # Output directory for extracted data
      - gitforai-cache:/cache  # Cache directory for embeddings/LLM responses
      - gitforai-vectordb:/vectordb  # Vector database persistent storage
      - gitforai-state:/home/gitforai/.gitforai  # State directory for incremental updates

    # Environment variables (override with .env file)
    environment:
      - LOG_LEVEL=${LOG_LEVEL:-INFO}
      - PYTHONUNBUFFERED=1
      # Embedding provider (default: local, no API key needed)
      - EMBEDDING_PROVIDER=${EMBEDDING_PROVIDER:-local}
      - EMBEDDING_MODEL=${EMBEDDING_MODEL:-all-MiniLM-L6-v2}
      # LLM API keys (optional - only required for openai provider)
      - OPENAI_API_KEY=${OPENAI_API_KEY}
      - ANTHROPIC_API_KEY=${ANTHROPIC_API_KEY}
      - LLM_MODEL=${LLM_MODEL:-gpt-4-turbo-preview}
      # Vector database configuration
      - VECTORDB_PROVIDER=${VECTORDB_PROVIDER:-chroma}
      - VECTORDB_PERSIST_DIR=${VECTORDB_PERSIST_DIR:-/vectordb}
      - VECTORDB_BATCH_SIZE=${VECTORDB_BATCH_SIZE:-100}

    # Override default command
    # Uncomment and modify as needed:
    # command: extract /repos/myrepo --output /output/commits.json

    # Resource limits (optional)
    deploy:
      resources:
        limits:
          cpus: '2.0'
          memory: 2G
        reservations:
          cpus: '0.5'
          memory: 512M

    # Network mode
    network_mode: bridge

    # Restart policy
    restart: "no"

# Named volumes for persistence
volumes:
  gitforai-cache:
    driver: local
  gitforai-vectordb:
    driver: local
  gitforai-state:
    driver: local

# Networks (for future multi-service setup)
networks:
  default:
    name: gitforai-network