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.
npx agentmods add skills/demodev-lab/claude-code-plugin-demokit/loopnpx skills add demodev-lab/claude-code-plugin-demokit --skill loopgit clone --depth 1 https://github.com/demodev-lab/claude-code-plugin-demokitWhat 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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00041 | $0.01002 |
| Opus 5 | $0.00020 | $0.00501 |
| Sonnet 5 | $0.00008 | $0.00200 |
| Haiku 4.5 | $0.00004 | $0.00100 |
Grade A, and why
loop 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.
How it starts
The opening of the file, as written. The whole thing — 123 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/loop - 자율 반복 루프
Ralph Wiggum 스타일: Stop hook으로 Claude 종료를 가로채서 같은 prompt를 재실행. 작업 완료까지 자동으로 반복하며, 매 반복마다 이전 결과를 확인하고 개선.
help
인자가 help이면 아래 도움말만 출력하고 실행을 중단한다:
/loop — 작업 완료까지 자동 반복 실행
사용법:
/loop {prompt} [--max-iterations N] [--completion-promise "..."]
파라미터:
prompt 실행할 작업 프롬프트 (필수)
옵션:
--max-iterations 최대 반복 횟수 (기본 10)
--completion-promise 완료 신호 문자열 (기본 LOOP_DONE)
예시:
/loop User 도메인 CRUD 구현 + 테스트 작성
/loop 전체 테스트 통과시켜줘 --max-iterations 30
/loop 인증 시스템 구현 --completion-promise "AUTH_COMPLETE"
관련 명령:
/cancel-loop — 루프 즉시 취소
/pdca iterate — 설계-구현 Gap 반복 개선
동작 원리
1. /loop 실행 → 루프 상태 활성화
2. Claude가 작업 수행
3. Claude가 멈추려 하면 Stop hook이 가로챔
4. 같은 prompt를 재주입 → Claude가 이전 결과를 보고 계속 작업
5. completion promise 감지 또는 max-iterations 도달 시 종료
실행 절차
1. 루프 활성화
사용자 prompt를 파싱하여 아래 명령을 Bash로 실행:
node scripts/loop-ctl.js start \
--prompt "{사용자가 지정한 작업 프롬프트}" \
--max-iterations {횟수, 기본 10} \
--completion-promise "{완료 신호, 기본 LOOP_DONE}"
2. 작업 시작
루프 활성화 후 즉시 prompt에 명시된 작업을 시작한다.
3. 매 반복마다 수행할 것
- 이전 반복에서 작성한 코드/파일 확인
- git diff 또는 파일 읽기로 현재 상태 파악
- 빌드/테스트 실행하여 문제 확인
- 발견된 문제 수정
- 새 기능 추가 (아직 남은 것이 있다면)
4. 종료 조건
아래 중 하나를 만족하면 루프가 종료된다:
- 완료 신호: 응답에 completion promise 문자열을 포함
- 기본값:
LOOP_DONE - 모든 작업이 완료되었을 때만 포함할 것
- 기본값:
- 최대 반복: max-iterations 도달 (안전장치)
- 수동 취소:
/cancel-loop실행
5. 완료 시
작업이 모두 끝나면:
- 최종 결과를 요약
- 응답에 completion promise 문자열을 포함 (예:
LOOP_DONE) - 루프가 자동으로 종료됨
사용 예시
# 기본 사용
/loop User 도메인 CRUD 구현 + 테스트 작성 + 빌드 통과시켜줘
# 반복 횟수 지정
/loop Order API 전체 구현 --max-iterations 30
# 완료 신호 커스텀
/loop 인증 시스템 구현 --completion-promise "AUTH_COMPLETE"
# 테스트 통과까지 반복
/loop 전체 테스트 통과시켜줘. 실패하는 테스트를 수정하고 ./gradlew test 통과할 때까지 반복
# 코드 리뷰 + 수정 반복
/loop 코드 리뷰 후 발견된 문제 모두 수정. 더 이상 수정할 것이 없을 때 종료
적합한 작업
적합:
- 명확한 완료 기준이 있는 작업 (테스트 통과, 빌드 성공)
- 반복적 개선이 필요한 작업 (코드 품질 개선, 리팩토링)
- 여러 파일에 걸친 대규모 구현
- 무인 실행이 가능한 작업 (밤새 실행)
부적합:
- 사람의 판단/디자인 결정이 필요한 작업
- 완료 기준이 모호한 작업
- 일회성 단순 작업
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- yesterday First seen · 123 lines · 41 tokens per session scan A 0da75a81a88a
loop is a skill published in the GitHub repository demodev-lab/claude-code-plugin-demokit (2 stars, last pushed 6mo ago), licensed MIT. It adds 41 tokens to every session and 1,002 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.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…