Automated summary
A guide for delegating information gathering (project code, examples, external documentation) to search subagents spawned through the Task tool instead of searching directly.
Strengths
- Clear routing table mapping each need (codebase, examples, web) to the right agent and model
- Demonstrates the parallel spawn pattern by issuing several Task calls in a single message
- Explicitly bans running the claude CLI through Bash and requires the Task tool
- Documents the follow-up flow: consolidate results, break down tasks, hand off to coder-agent
Limitations
- Assumes .claude/agents/codebase-search-agent.md, reference-agent.md and web-search-agent.md already exist in the project
- Contains no search logic itself — it is dispatch guidance, so quality depends on the target agents
- Claude Code only, and the 500-character summary cap can drop useful detail
Use it when you need to gather repository context, existing examples or external docs quickly before planning or writing code.
Skip it when the referenced search subagents are not defined in the project, or for a trivial file lookup that Glob/Grep/Read already covers.
Security analysis
SafeThe skill is a documentation guide that constrains the model to Task/Read/Glob/Grep and explicitly forbids using Bash or invoking the claude CLI. It contains no destructive commands, credential exfiltration, obfuscated payloads, or safety-disabling instructions. Subagent prompts are limited to read-only search and summarization.
No concerns found
How this analysis is madeExamples
Find all authentication-related code in this project — login, session and token handling — and summarize the files and functions involved before we plan the refactor.Search the official React 18 documentation for what changed in useEffect compared to React 17, and summarize the key differences.Look through examples/ and samples/ for existing API endpoint implementations so we can follow the same convention.name: spawn-search-agents description: 정보 수집이 필요할 때 검색 에이전트 활용 가이드. 코드 탐색, 레퍼런스, 웹검색 allowed-tools: Task, Read, Glob, Grep
검색 에이전트 활용 가이드
정보 수집이 필요할 때 적절한 검색 에이전트를 선택하고 활용하는 방법입니다.
언제 사용하나?
- 코드 구조 파악이 필요할 때
- 기존 패턴/컨벤션 확인이 필요할 때
- 외부 라이브러리 문서가 필요할 때
- 유사 구현 예제가 필요할 때
중요: /orchestrate는 직접 검색 금지
/orchestrate 커맨드는 Glob, Grep, Search를 직접 사용하면 안 됩니다. 반드시 아래 검색 에이전트를 Task tool로 spawn해야 합니다.
절대 금지: Bash로 claude 명령어 실행
- ❌
claude --agent {agent-name}사용 금지 - ❌
claude task --subagent {agent-name}사용 금지 - ❌ Bash tool로 claude CLI 실행 금지
반드시 Task tool (function call)을 사용하세요.
에이전트별 역할
| 에이전트 | 용도 | 모델 | |---------|------|------| | codebase-search-agent | 프로젝트 내 코드 탐색 | Sonnet | | reference-agent | 예제/템플릿 코드 탐색 | Haiku | | web-search-agent | 외부 문서/API 레퍼런스 | Haiku |
중요: 반드시 Task tool을 사용합니다. Bash 명령어로 실행하지 마세요.
codebase-search-agent
용도:
- 관련 파일, 함수, 클래스 찾기
- 의존성 관계 분석
- 기존 코드 패턴 파악
- 코드 컨벤션 확인
Spawn 예시:
Task tool 호출:
subagent_type: "codebase-search-agent"
run_in_background: true
prompt: |
사용자 인증 관련 코드를 찾아줘. 로그인, 세션, 토큰 관련 파일과 함수를 파악해줘.
.claude/agents/codebase-search-agent.md 의 지시를 따르고,
작업 완료 후 결과만 요약해서 보고해줘.
reference-agent
용도:
- 프로젝트 내 예제 코드 찾기
- 템플릿 파일 탐색
- 유사 구현 사례 발굴
Spawn 예시:
Task tool 호출:
subagent_type: "reference-agent"
run_in_background: true
prompt: |
API 엔드포인트 구현 예제를 찾아줘. examples/, samples/ 폴더와 README를 확인해줘.
.claude/agents/reference-agent.md 의 지시를 따르고,
작업 완료 후 결과만 요약해서 보고해줘.
web-search-agent
용도:
- 공식 문서 검색
- API 레퍼런스 조회
- 베스트 프랙티스 검색
- 에러 해결책 검색
Spawn 예시:
Task tool 호출:
subagent_type: "web-search-agent"
run_in_background: true
prompt: |
React 18의 useEffect 변경사항에 대해 검색해줘. 공식 문서 위주로 찾아줘.
.claude/agents/web-search-agent.md 의 지시를 따르고,
작업 완료 후 결과만 요약해서 보고해줘.
병렬 Spawn 패턴
시간 단축을 위해 동시에 Task tool 호출 (단일 메시지에 여러 호출):
Task tool #1:
subagent_type: "codebase-search-agent"
run_in_background: true
prompt: "프로젝트 내 관련 코드 탐색..."
Task tool #2:
subagent_type: "reference-agent"
run_in_background: true
prompt: "예제/템플릿 탐색..."
Task tool #3 (필요시):
subagent_type: "web-search-agent"
run_in_background: true
prompt: "외부 문서 검색..."
결과 활용
정보 수집 완료 후
- 각 에이전트 결과 취합
- todo-list-agent로 작업 분해
- task 파일 생성
- coder-agent에 정보 전달
결과가 불충분할 때
- 키워드 변경하여 재탐색
- 다른 에이전트 추가 spawn
- 사용자에게 추가 정보 요청
주의사항
- 검색 에이전트는 정보 수집만 담당
- 코드 작성/수정은 coder-agent에게 위임
- 결과는 500자 이내 요약으로 받음
<!-- SKILL-PROJECT-CONFIG-START --> <!-- 프로젝트 특화 설정이 /orchestration-init에 의해 이 위치에 추가됩니다 --> <!-- SKILL-PROJECT-CONFIG-END -->
From the file to a first result
The download contains exactly the content displayed on this page. This copy may differ from the latest version in the source repository.
View the source and supporting filesThis download contains SKILL.md only. Any scripts, references, templates and dependencies are not included. Get the required files from the source, preserving their directory structure, before continuing.
1. Download the file
Download SKILL.mdFrom the root of a test project, save the file as .skill-downloads/spawn-search-agents-mmbizjo2/SKILL.md, or use the command below. This folder is for inspecting the skill before installation.
Command for macOS/Linux or WSL: requires a POSIX shell and curl. On Windows without WSL, use the download button and create the folder manually.
mkdir -p '.skill-downloads/spawn-search-agents-mmbizjo2' && curl --fail --show-error --location 'https://skills-guides.com/api/skills/spawn-search-agents-mmbizjo2/download' --output '.skill-downloads/spawn-search-agents-mmbizjo2/SKILL.md.part' && mv '.skill-downloads/spawn-search-agents-mmbizjo2/SKILL.md.part' '.skill-downloads/spawn-search-agents-mmbizjo2/SKILL.md'2. Check prerequisites
Read SKILL.md before using it. If it references scripts or external services, inspect them and prepare the required access in your test environment.
Tools declared by the author (check availability in your agent): Task, Read, Glob, Grep
No recognized prerequisites section in this file. This does not mean it is self-contained: libraries, tools and access requirements may be mentioned elsewhere in the content or source repository.
The platforms on this page are catalog indications. The paths below follow tool documentation; they do not prove this skill has been run in those tools. No minimum tool version is confirmed for this listing.
3. Load the skill
After inspection, copy the complete folder into your project using one of the commands below, then open a new tool session. The command stops if a folder with the same name already exists: compare the files before replacing it.
Claude Code
test -f '.skill-downloads/spawn-search-agents-mmbizjo2/SKILL.md' && test -s '.skill-downloads/spawn-search-agents-mmbizjo2/SKILL.md' && mkdir -p '.claude/skills' && test ! -e '.claude/skills/spawn-search-agents' && cp -R '.skill-downloads/spawn-search-agents-mmbizjo2' '.claude/skills/spawn-search-agents'OpenSkills
OpenSkills must already be installed. The local install/read procedure is checked with OpenSkills 1.5.0, which requires Node.js ≥ 20.6.0. Catalog slugs are not resolved automatically.
Official OpenSkills installationopenskills install './.skill-downloads/spawn-search-agents-mmbizjo2'openskills read 'spawn-search-agents-mmbizjo2'Reading should display the installed file content and its base directory. It does not run the skill or install its dependencies.
For an agent that reads AGENTS.md, then sync the index. This command changes AGENTS.md: review the diff before using it.
openskills syncManual method or another tool: attach SKILL.md to a conversation, or open it in an agent that can read files. Include any required supporting resources. No universal folder is recognized by every tool.
4. Try a first task
Use a short, synthetic input: an API endpoint, a small document or an interface brief. In a new conversation, attach the file or give the agent its path, then adapt this request:
Read .skill-downloads/spawn-search-agents-mmbizjo2/SKILL.md and apply its instructions to this example: [describe a test input]. Identify any missing files, dependencies or access. Produce the result requested by the skill without inventing data or running unreviewed scripts.Expected result: output matching the skill instructions, or a precise list of missing prerequisites. A read confirmation alone does not validate a first use. If an external resource is unavailable, that path remains untested.
React Hook Form Best Practices
Development
Master React Hook Form: avoid anti-patterns, manage subscriptions, and handle reset, dirty state, and controlled inputs correctly.
Emotion to Mantine Migration
Development
Migrate Emotion styled-components to Mantine components with style props and CSS modules. Use when converting .styled.tsx files or removing @emotion imports.
Scope Creep Detector
Development
Analyzes git diffs against a stated intent to detect scope creep, unrelated files, broad PRs, and oversized changes, then suggests keep, split, or justify actions.