Guide for using Claude Code subagents effectively. Use when deciding whether to delegate a task to a subagent or handle it directly.
Delegates well-defined coding tasks like multi-file implementations, refactoring, or bug fixes to autonomous subagents. Triggers when you mention specific code changes with clear requirements and acceptance criteria.
/plugin marketplace add seajhawk/cc-marketplace/plugin install auto-dev@cc-marketplaceThis skill inherits all available tools. When active, it can use any tool Claude has access to.
Subagents are autonomous Claude instances that work independently on well-defined tasks. They have their own context, tools, and completion criteria.
coder SubagentPurpose: Autonomous coding agent that implements features and fixes bugs
Tools: Read, Write, Edit, Bash, Grep, Glob
Workflow:
Test Gate: When the coder completes, the SubagentStop hook runs tests. If tests fail, the subagent is blocked and must fix issues before completion.
The UserPromptSubmit hook auto-classifies prompts. Coding tasks are automatically suggested for subagent delegation.
Prefix your prompt with subagent: to force delegation:
subagent: implement user authentication
For multiple independent coding tasks, subagents can run in parallel:
1. Add user profile page
2. Implement password reset
3. Create admin dashboard
The classifier will detect these as separate coding tasks and suggest parallel execution.
subagent: Add search functionality to the products page
Requirements:
- Search input with debounced API calls
- Display results in existing product grid
- Handle empty state and loading state
- Add tests for search logic
Files: src/pages/Products.tsx, src/api/products.ts
Fix the issue where form validation doesn't trigger on blur
Steps to reproduce:
1. Go to contact form
2. Click into email field, then click out
3. No validation error appears (should show "Email required")
Related: src/components/ContactForm.tsx
This skill should be used when the user asks to "create a slash command", "add a command", "write a custom command", "define command arguments", "use command frontmatter", "organize commands", "create command with file references", "interactive command", "use AskUserQuestion in command", or needs guidance on slash command structure, YAML frontmatter fields, dynamic arguments, bash execution in commands, user interaction patterns, or command development best practices for Claude Code.
This skill should be used when the user asks to "create an agent", "add an agent", "write a subagent", "agent frontmatter", "when to use description", "agent examples", "agent tools", "agent colors", "autonomous agent", or needs guidance on agent structure, system prompts, triggering conditions, or agent development best practices for Claude Code plugins.
This skill should be used when the user asks to "create a hook", "add a PreToolUse/PostToolUse/Stop hook", "validate tool use", "implement prompt-based hooks", "use ${CLAUDE_PLUGIN_ROOT}", "set up event-driven automation", "block dangerous commands", or mentions hook events (PreToolUse, PostToolUse, Stop, SubagentStop, SessionStart, SessionEnd, UserPromptSubmit, PreCompact, Notification). Provides comprehensive guidance for creating and implementing Claude Code plugin hooks with focus on advanced prompt-based hooks API.