PrepGenAICerts
Sign in
Courses/Introduction to Subagents

Introduction to Subagents

Course

A comprehensive, hands-on course on Claude subagents — specialized AI assistants that run in their own context window, do focused work, and return only a clean summary. Learn how subagents preserve context, when to delegate (and when not to), how to create custom subagents in Claude Code (.claude/agents) and the Claude Agent SDK (AgentDefinition), the built-in Explore/Plan/general-purpose agents, tool restriction, model selection, parallel vs sequential execution, and the orchestrator-worker patterns behind Anthropic's multi-agent research system. Verified against official Anthropic docs. Free to start.

5 sections|18 lessons|0/18 completed
1What Are Subagents?2Why Context Isolation Matters3The Built-in Subagents: Explore, Plan, General-Purpose4How Delegation Works: The Description Drives Everything
5Creating Your First Subagent with /agents6The Config File: Frontmatter and System Prompt7Scope and Precedence: Where Subagents Live8Choosing Model, Tools, and Color
9Descriptions That Shape Input Prompts10Output Formats and Obstacle Reporting11Limiting Tool Access
12When Subagents Shine13Anti-Patterns: When Subagents Hurt14The Decision Rule and Execution Modes
15Subagents in the Claude Agent SDK16Parallel, Sequential, and the Workflow Tool17Multi-Agent Orchestration: The Orchestrator-Worker Pattern18Course Review and Exam Tips

PrepGenAICerts is an independent exam-prep site. It is not affiliated with, endorsed by, or sponsored by Anthropic, PBC. Claude and Anthropic are trademarks of Anthropic, PBC. Exam details are taken from publicly available Anthropic and Pearson documentation. About PrepGenAICerts.