새로운 Claude Code subagent를 생성하는 agent.
/plugin marketplace add wonjiko/claude-toymarket/plugin install matryoshka-plugin@claude-toymarket새로운 Claude Code subagent를 생성하는 agent.
~/.claude/plugins/marketplaces/claude-plugins-official/plugins/*/agents/ 에서 기존 agent 예시 확인agents/[name].md 생성---
name: [agent-name]
description: Use this agent when [트리거 조건]...
<example>
Context: [상황]
user: "[사용자 메시지]"
assistant: "[응답]"
<commentary>
[왜 이 agent가 트리거되어야 하는지]
</commentary>
</example>
model: sonnet
tools: ["Read", "Write", ...]
---
[System prompt - agent의 역할, 책임, 프로세스 정의]
| 필드 | 설명 |
|---|---|
| name | agent 식별자 (lowercase, hyphens) |
| description | 트리거 조건 ("Use this agent when...") |
| model | 사용할 모델 (sonnet, opus, haiku) |
| tools | 사용 가능한 도구 목록 |
생성된 agent 파일 경로와 내용 요약을 출력한다.
Use this agent when analyzing conversation transcripts to find behaviors worth preventing with hooks. Examples: <example>Context: User is running /hookify command without arguments user: "/hookify" assistant: "I'll analyze the conversation to find behaviors you want to prevent" <commentary>The /hookify command without arguments triggers conversation analysis to find unwanted behaviors.</commentary></example><example>Context: User wants to create hooks from recent frustrations user: "Can you look back at this conversation and help me create hooks for the mistakes you made?" assistant: "I'll use the conversation-analyzer agent to identify the issues and suggest hooks." <commentary>User explicitly asks to analyze conversation for mistakes that should be prevented.</commentary></example>