← Back to Glossary
Multi-Agent System
An architecture where multiple specialized LLM agents collaborate, share state, and delegate sub-tasks to accomplish complex goals.
Why it exists
A single agent with dozens of tools often misfires. Specialized agents (e.g. Coder, Tester, Reviewer) perform better in focused roles.
How it works
Agents communicate via supervisor routing or message passing protocols, each operating with dedicated prompts and tools.