# docker/compose-for-agents

Build and run AI agents using Docker Compose. A collection of ready-to-use examples for orchestrating open-source LLMs, tools, and agent runtimes.

Repository: https://github.com/docker/compose-for-agents
Canonical: https://ross.abutalabs.com/products/compose-for-agents
Language: TypeScript
License: Apache-2.0
License Family: permissive
Topics: agentic-workflows, ai-agents, docker, docker-compose, examples, large-language-models, llama-cpp, openai-gym, self-hosted
Last push: 2026-09-02T19:09:39+00:00

## Health v2 (maintenance only)
Score: 64/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 100, release rhythm 35, longevity 34
- inputs: {"age_days": 481, "days_push": 0, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1028, forks 468 (observed 2026-09-03T02:15:11.329758+00:00)

## What it is
A collection of ready-to-use Docker Compose examples for building and running AI agents with open-source LLMs, MCP tools, and agent runtimes like CrewAI, LangGraph, and Google ADK. Each demo is self-contained and can run models locally via Docker Model Runner or with OpenAI's API.

## Use cases
- run ai agents locally with docker compose
- orchestrate multi-agent workflows with docker
- self-host open-source llm agents
- learn how to wire mcp servers into agent stacks
- run local models like qwen3 with docker model runner
- example compose files for langgraph or crewai agents

## When to choose
- you want working, copy-paste examples of agent stacks on Docker Compose
- you want to run open models locally on a GPU machine
- you want to experiment with MCP tools alongside agent frameworks

## When to avoid
- you need a production agent platform rather than demos
- you don't use Docker or lack GPU/remote compute
- you need a framework library to import into your own code

## Facets
- artifact type: learning-resource
- maturity: active
- function: agent-framework, mcp, llm-inference, container-orchestration, developer-tools
- domain: large-language-models, developer-tools, self-hosted
- platform: cross-platform
- tags: docker-compose, ai-agents, mcp, local-llm, examples, demos, model-runner, docker, gpu

## Member repositories
- docker/compose-for-agents (main) score 64

## Provenance
- Observed fields: from GitHub, fetched 2026-09-03T02:15:11.329758+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T07:08:03.339045+00:00, confidence not recorded.
  - readme: https://github.com/docker/compose-for-agents (fetched 2026-09-03T02:15:11.329758+00:00, sha 6ee579704adf)
- Data as of 2026-08-30T08:39:29.467469+00:00.