Claude Code now supports orchestrating multiple AI agent instances working together as a team. Agent teams enable parallel work through independent Claude sessions that share a task list, communicate via messaging, and coordinate through a lead agent. Best suited for research, code review, debugging with competing hypotheses,
Table of contents
When to use agent teams Enable agent teams Start your first agent team Control your agent team How agent teams work Use case examples Best practices Troubleshooting Limitations Next stepsSort: