# Multi-Agent System

> An architecture where multiple specialized agents collaborate through messages, tools, or a shared controller.

- Category: Agents & Autonomy
- Stage: Rising
- Also called: MAS, Agent collective
- Canonical: https://www.joelcsaji.com/ai-dictionary/multi-agent-system

## Plain English

A team of narrow agents can be easier to steer than one giant all-purpose agent.

## Example

A planning agent delegates to a browser agent, a coding agent, and a QA agent.

## Why it matters

Multi-agent patterns are becoming common when context grows too large or expertise needs to be separated.

## Related

- [Agent Swarm](https://www.joelcsaji.com/ai-dictionary/agent-swarm)
- [Handoff](https://www.joelcsaji.com/ai-dictionary/handoff)
- [Supervisor Agent](https://www.joelcsaji.com/ai-dictionary/supervisor-agent)

## Sources

- [LangChain: Multi-Agent](https://docs.langchain.com/oss/python/langchain/multi-agent)
- [AutoGen: Multi-Agent Chat](https://autogenhub.github.io/autogen/docs/Use-Cases/agent_chat/)
- [Google ADK: A2A](https://adk.dev/a2a/intro/)