agent-creator

A guide for creating task-specific Claude Code sub-agents, which are smaller assistants configured for particular jobs. It covers their instructions, model choice, allowed tools, permissions, files, and validation.

In plain words
What is it for?
Use it to create agents for tasks such as code implementation, analysis, review, exploration, or verification.
Why use it?
It turns a broad request for a custom assistant into a defined configuration that can be reused in a project or plugin.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/this-hw/claude-code-kit/agent-creator
Any agent
npx skills add This-HW/claude-code-kit --skill agent-creator
Clone the repo
git clone --depth 1 https://github.com/This-HW/claude-code-kit

Made for: Claude Code, Codex.

Per session 39 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 725 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00039 $0.00725
Opus 5 $0.00019 $0.00362
Sonnet 5 $0.00008 $0.00145
Haiku 4.5 $0.00004 $0.00072

Measured yesterday against content hash 9e92f30d64b6, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

agent-creator scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugins/common/skills/agent-creator/SKILL.md · 113 lines

What it actually says

Agent Creator

새로운 Claude Code 서브에이전트를 생성하고 최적화합니다.

생성 워크플로우

  1. 요구사항 수집

    • 에이전트의 목적
    • 사용 시점 (자동 위임 트리거)
    • 필요한 도구
    • 파일 수정 필요 여부
  2. 설정 결정

    • 모델 선택 (Opus/Sonnet/Haiku)
    • 권한 모드 선택
    • 도구 화이트리스트/블랙리스트
  3. 파일 생성

    • 위치: plugins/{domain}/agents/{category}/ (플러그인 구조)
    • 또는 .claude/agents/ (프로젝트 로컬)
    • 파일명: agent-name.md
  4. 검증

    • frontmatter 형식 확인
    • 도구 이름 검증
    • plugin.json 등록 확인

모델 선택 가이드

작업 유형 권장 모델 이유
전략/분석/리뷰 opus 복잡한 추론
코드 구현/수정 sonnet 균형잡힌 성능
탐색/검증/단순작업 haiku 빠른 실행, 비용 효율

권한 모드 선택

모드 용도
default 표준 권한 확인
acceptEdits 파일 편집 자동 허용
plan 읽기 전용

출력 형식 (claude-code-kit 플러그인 구조)

---
name: [소문자-하이픈]
description: [역할]. MUST USE when: [트리거 조건].
model: [sonnet|opus|haiku]
effort: [low|medium|high|max]
tools:
  - [도구 목록]
disallowedTools:
  - Task
---

[시스템 프롬프트]

플러그인 구조 등록

생성 후 plugins/{domain}/.claude-plugin/plugin.jsonagents 필드에 등록:

{
  "agents": "agents/"
}

에이전트 파일은 plugins/{domain}/agents/{category}/{name}.md에 위치.

자동 위임 최적화

description에 포함하면 자동 위임 빈도 증가:

  • "MUST USE when:"
  • "Use PROACTIVELY"
  • "Use immediately after"

예제

---
name: test-runner
description: Test execution specialist. MUST USE when: code changes are made and tests need to be run.
model: haiku
effort: low
tools:
  - Read
  - Bash
  - Grep
  - Glob
disallowedTools:
  - Task
---

You are a test automation expert.

When invoked:

1. Identify changed files
2. Run relevant tests
3. Report results with fix suggestions if failed
Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday First seen · 113 lines · 39 tokens per session scan A 9e92f30d64b6

Subscribe to this mod's changes

agent-creator is a skill published in the GitHub repository This-HW/claude-code-kit (4 stars, last pushed 4d ago), licensed MIT. It adds 39 tokens to every session and 725 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories

watch

File sentinel that monitors the working directory for changes and marker comments, then auto-triggers appropriate skills. Poll-based via git diff against the last scan commit. Writes intake items for batch processing and routes marker actions through /do. Use for automatic reactions to file changes; do NOT use for…

SethGammon/Citadel · 70 tokens

pr-watch

Local PR watcher. Monitors CI status, automatically fixes failing checks by reading failure logs and applying targeted fixes, then optionally merges when all checks pass. Local CLI analog to Claude Code's cloud auto-fix feature.

SethGammon/Citadel · 46 tokens

qa

Browser-based QA verification. Launches a real browser, navigates the app, clicks buttons, fills forms, and tests user flows. Works as a standalone skill or as a phase end condition in campaigns. Requires Playwright (optional dependency, graceful skip if not installed).

SethGammon/Citadel · 56 tokens

review

5-pass structured code review — correctness, security, performance, readability, consistency.

SethGammon/Citadel · 17 tokens

live-preview

Mid-build visual verification loop. Takes screenshots of components during construction, not just after. Catches visual regressions and invisible features before they compound. Requires Playwright or similar screenshot tool.

SethGammon/Citadel · 40 tokens

wiki

Markdown-first knowledge base where the LLM acts as librarian. Ingests raw sources, compiles and interlinks topic files, self-maintains an index. No vector DB or embeddings required -- uses LLM-native navigation over structured markdown up to 400K words.

SethGammon/Citadel · 56 tokens