Skip to main content
The service catalog is the foundation of better-openclaw. Each service is a containerized application that you can add to your stack with automatic dependency resolution, port management, and configuration.

​
What is a service?

A service in better-openclaw is a pre-configured Docker container with:
  • Pinned image tags for reproducibility (e.g., postgres:17-alpine, redis:8-alpine)
  • Port mappings with automatic conflict detection
  • Volume definitions for persistent data
  • Environment variables with secure default generation
  • Health checks for container readiness
  • Dependency declarations that trigger automatic service installation
  • Skills that integrate with OpenClaw agents

​
Service categories

The 94+ services are organized into 21 categories:
Claude Code, Codex, Gemini CLI, Kimi, OpenCode
AnythingLLM, Dify, Flowise, LibreChat, LiteLLM, Open WebUI
ComfyUI, Ollama, Stable Diffusion, Whisper
Cal.com, Home Assistant, n8n, Temporal, xyOps
Convex, Neo4j, PostgreSQL, Redis, Supabase, Valkey
ChromaDB, Milvus, Qdrant, Weaviate
FFmpeg, Immich, Jellyfin, Motion Canvas, Remotion
Ghost, Mixpost, Postiz
Matomo, OpenPanel, Umami
AppFlowy, DocsGPT, NocoDB, Outline, Paperless-ngx
MinIO, Nextcloud
Beszel, Code Server, Convex Dashboard, Coolify, Dokploy, Dozzle, Gitea, Headscale, Jenkins, Mission Control, Portainer, Tailscale, Watchtower
Caddy, Traefik
Grafana, Loki, Prometheus, SigNoz, Uptime Kuma
Browserless, LightPanda, Playwright Server, Scrapling, Steel Browser
Gotify, La Suite Meet (frontend/backend/agents), LiveKit, Matrix Synapse, Mattermost, ntfy, Rocket.Chat, UseSend
Authentik, CrowdSec, HexStrike, PentAGI, PentestAgent, SolidityGuard, Vaultwarden
Stream Gateway
Desktop Environment (KasmVNC)

​
Service definition structure

Each service is defined with a TypeScript schema that includes: 
export const redisDefinition: ServiceDefinition = {
  id: "redis",
  name: "Redis",
  description: "In-memory data store for caching, sessions, and messaging",
  category: "database",
  icon: "🔴",

  image: "redis",
  imageTag: "8-alpine",
  ports: [{
    host: 6379,
    container: 6379,
    description: "Redis server port",
    exposed: true,
  }],
  volumes: [{
    name: "redis-data",
    containerPath: "/data",
    description: "Persistent Redis data",
  }],
  environment: [{
    key: "REDIS_PASSWORD",
    defaultValue: "changeme",
    secret: true,
    description: "Password for Redis authentication",
    required: true,
  }],
  healthcheck: {
    test: "redis-cli ping",
    interval: "10s",
    timeout: "5s",
    retries: 3,
  },

  skills: [{ skillId: "redis-cache", autoInstall: true }],
  requires: [],
  recommends: [],
  conflictsWith: ["valkey"],

  minMemoryMB: 128,
  gpuRequired: false,
};

​
Dependency resolution

When you add a service, better-openclaw automatically resolves dependencies. For example:
1

Select n8n

You choose to add the n8n workflow automation service
2

Dependency detected

n8n declares requires: ["postgresql"] in its definition
3

PostgreSQL added

PostgreSQL is automatically added to your stack
4

Configuration wired

n8n’s environment variables are automatically configured to connect to PostgreSQL

​
Service conflicts

Some services conflict with each other. For example, Redis and Valkey both implement the Redis protocol and use port 6379. If you try to add both, better-openclaw will warn you and prevent the conflict.

​
Resource requirements

Each service declares its minimum memory requirement:
  • Redis: 128 MB
  • PostgreSQL: 256 MB
  • n8n: 512 MB
  • Ollama: 4 GB (8 GB recommended with models)
The generator calculates total memory requirements and warns you if your stack exceeds available resources.

​
Listing services

You can browse the full service catalog with: 
npx create-better-openclaw services list

​
Service metadata

Each service includes:
  • Tags: Searchable keywords (e.g., cache, workflow, vector-db)
  • Maturity: stable, beta, or experimental
  • Documentation URL: Official docs link
  • Native support: Whether the service can run natively on bare metal

​
Skills integration

Many services include OpenClaw skills that are automatically installed. For example:
  • Redis → redis-cache skill for caching operations
  • Qdrant → qdrant-memory skill for vector search
  • n8n → n8n-trigger skill for workflow automation
  • FFmpeg → ffmpeg-process skill for video processing
These skills provide ready-to-use templates for AI agents to interact with services.

Related concepts

Learn how skills are bundled into skill packs

Build docs developers (and LLMs) love

Get started for freeTalk to us