creating-subagents
Creating Subagents
This Skill provides comprehensive knowledge about creating effective subagents in Claude Code.
What Are Subagents?
Subagents are specialized AI assistants that Claude Code can delegate tasks to. Each subagent:
- Has a specific purpose and expertise area
- Uses its own context window separate from the main conversation
- Can be configured with specific tools it's allowed to use
- Includes a custom system prompt that guides its behavior
Critical Insight - The Primary Value: While specialization is important, the primary value of subagents is token context isolation. Subagents allow the main conversation to delegate "token-heavy" or "noisy" tasks (like extensive research, analyzing large codebases, or processing logs) to a separate context window. This keeps the main orchestrator's context clean and focused, enabling much longer-running and more complex workflows.
Think of subagents as allowing the main agent to "go on side quests" without polluting its main context with unnecessary details.
Mental Model: Orchestrator and Specialists
The most effective way to think about subagents is as a distributed team:
More from mikkelkrogsholm/wp-plugins
wordpress-core
Core WordPress plugin development fundamentals including file structure, security patterns (sanitize, escape, nonces, capabilities), hooks system (actions/filters), database operations with wpdb, and coding standards. Use when creating plugins, implementing security, or working with core WordPress APIs.
3creating-commands
Expert knowledge on creating slash commands for Claude Code. Use when designing or creating slash command .md files, understanding command arguments, or implementing bash execution.
1