Built-in Subagents - Claude Code
Researcher, auditor, reviewer, and other ready-made subagent types.
Claude Code ships with a handful of pre-built subagent types so you can delegate without writing your own first.
What it does
Built-in subagents include roles like researcher (gathers context without editing), auditor (reviews code for issues), and reviewer (checks changes). Each has a tuned system prompt and tool allowlist. You invoke them by name or let the main agent pick the right one for the task.
When to use it
- Starting point before you build custom subagent definitions.
- Situations where you just need a clear role label, not a bespoke prompt.
- Team onboarding - everyone has the same built-in types available.
- Quick delegation without the overhead of writing a new agent.
Gotchas
- Built-ins are generic. For repeat workflows, a custom subagent usually wins.
- The list and behavior of built-ins evolves between releases. Don't hard-code assumptions.
- You can shadow a built-in with a same-named custom agent - useful or confusing depending on context.
Official docs: https://code.claude.com/docs/en/sub-agents.md#built-in-subagents
Technical content at the intersection of AI and development. Building with AI agents, Claude Code, and modern dev tools - then showing you exactly how it works.
Get the weekly deep dive
Tutorials on Claude Code, AI agents, and dev tools - delivered free every week.
Was this helpful?
