공식 출처부터 커뮤니티까지 전체 Tier(1-4)를 조사하고 비교 분석하는 전략입니다.
Conducts comprehensive multi-tier research from official sources to community forums for reliable technical answers.
/plugin marketplace add inchan/cc-plugins/plugin install inchan-tdd@inchan-claude-plugin공식 출처부터 커뮤니티까지 전체 Tier(1-4)를 조사하고 비교 분석하는 전략입니다.
허용: Tier 1-4 (모두) 다양성 보장: 각 Tier별 최소 2개 결과 수집
*.org, docs.*, *.dev총 목표: 최소 10개 결과
// Tier별 쿼리 생성
tier1_2_query = "{쿼리} site:{공식도메인}"
tier3_query = "{쿼리} site:stackoverflow.com OR site:developer.mozilla.org"
tier4_query = "{쿼리} tutorial OR guide OR example"
const [tier1, tier2, tier3, tier4] = await Promise.all([
WebSearch({ query: tier1_query, allowed_domains: tier1_domains }),
WebSearch({ query: tier2_query, allowed_domains: tier2_domains }),
WebSearch({ query: tier3_query, allowed_domains: tier3_domains }),
WebSearch({ query: tier4_query })
]);
IF tier1.length >= 2 AND tier2.length >= 2:
RETURN tier1 + tier2 // 공식 출처로 충분
ELSE:
CONTINUE to tier3 and tier4
total_score = domain_score + content_score + community_score + relevance_score
IF total_score >= 60:
RETURN result
ELSE:
FILTER OUT
diversity_check = {
"tier1_count": len([r for r in results if r.tier == 1]),
"tier2_count": len([r for r in results if r.tier == 2]),
"tier3_count": len([r for r in results if r.tier == 3]),
"tier4_count": len([r for r in results if r.tier == 4])
}
IF tier1_count < 2 OR tier2_count < 2:
WARNING("공식 출처 부족 - 추가 검색 필요")
| Tier | 최대 결과 | 이유 |
|---|---|---|
| Tier 1 | 5개 | 공식 관점 충분 |
| Tier 2 | 5개 | 실전 예제 충분 |
| Tier 3 | 7개 | 다양한 해결법 |
| Tier 4 | 3개 | 참고용 |
총 최대: 20개
참고: skills/search-core/resources/output-formats.md#2
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>