Borrowing it
Nothing to install: this file belongs to claude-code-expert/carve-harness. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/claude-code-expert/carve-harness/main/.claude/skills/checklist-loop/SKILL.mdgit clone --depth 1 https://github.com/claude-code-expert/carve-harnessWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/claude-code-expert/carve-harness/checklist-loop)<a href="https://agentmods.dev/skills/claude-code-expert/carve-harness/checklist-loop"><img src="https://agentmods.dev/badge/skills/claude-code-expert/carve-harness/checklist-loop/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/claude-code-expert/carve-harness/checklist-loop"><img src="https://agentmods.dev/badge/skills/claude-code-expert/carve-harness/checklist-loop.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Tool Misuse · line 81 Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).Fix: Validate all tool parameters against an allowlist. Reject dangerous parameter values (shell=True, --force, -rf /) and use safe defaults.
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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00083 | $0.02444 |
| Opus 5.5 | $0.00033 | $0.00978 |
| Sonnet 5.5 | $0.00017 | $0.00489 |
| Haiku 4.5 | $0.00008 | $0.00244 |
Grade A, and why
checklist-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 1mo ago.
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 — 109 lines — stays where its author put it; the contents beside it link to each section on GitHub.
checklist-loop — 스펙→개발→체크리스트→채점→재작업 루프 (모델 무관 SOP)
권위는 도구가 아니라 이 절차에 있다(fable-team-guide 4절).
carve-verify-loop워크플로는 이 절차를 코드로 자동 실행할 뿐이다. Fable/Opus/Sonnet 어떤 세션이든, 훅 없는 다른 에이전트든 아래를 그대로 수동 실행하면 동일하게 동작한다.
핵심 불변식
- 생성자(builder)와 채점자(evaluator)는 절대 같은 에이전트가 아니다 (Self-Eval Blindspot 방지, AGENTS.md 6절).
- 주장을 믿지 않는다. "구현했다"는 claim은 실제 코드 대조 + 테스트 실행으로만 인정한다.
- 완료 = 모든 항목 score ≥ 임계(기본 95). 하나라도 미달이면 루프는 끝나지 않는다.
- 에이전트 간 통신은 파일 기반 —
specs/checklist.json이 단일 진실원.
specs/checklist.json 스키마 (정본)
{
"goal": "주문 취소 API",
"iteration": 2,
"threshold": 95,
"items": [
{
"id": "c1",
"claim": "POST /orders/{id}/cancel 구현",
"acceptance": "204 반환 + 상태 CANCELLED 전이 테스트 통과",
"owns": ["src/api/cancel/**"],
"attempts": 2,
"score": 75,
"axes": { "exists": 25, "match": 25, "test": 0, "contract": 15, "no_regress": 10 },
"tests": { "ran": true, "passed": 3, "failed": 1, "command": "npm test -- cancel" },
"pass": false,
"gaps": ["멱등성 미검증", "이미 취소된 주문 409 테스트 없음"],
"evidence": "src/api/cancel/route.ts:12; test 3/4 pass"
}
]
}
claim/acceptance/owns: 분해 단계에서 채운다.ownsglob은 항목끼리 겹치면 안 된다(파일 오너 1개).type(선택):convention|correctness|domain_safety.domain_safety항목은 100점이 아니면 총점·임계와 무관하게 Stop 게이트가 차단한다(GATE-C7, 블루프린트 §5.5 허용 실패율 0%). 도메인 불변식(CLAUDE.md도메인 규칙)을 구현하는 항목에 붙여라. type 없는 항목은 임계(95) 규칙만.axes/tests/score/pass/gaps/evidence/attempts: 채점 단계에서 채운다.score = 5축 합,pass = score >= threshold.- 5축 루브릭(합 100):
exists25·match25·test25·contract15·no_regress10. test축은 채점자가 신고하지 않는다 — 실행 결과에서 파생한다. evaluator는tests: {ran, passed, failed, command, output}을 보고하고,ran:true·passed>0·failed=0이면test=25, 아니면 0이다(비율 부분점수 없음 — 4/5 통과가 20점이면 나머지 만점과 합쳐 정확히 95라 실패 테스트를 안고 통과한다). 미실행·실패 1건이면 나머지 네 축 만점이어도 합 75라 95를 넘을 수 없다(거짓 완료 차단). 숫자를 직접 받으면 이 불변식은 산문일 뿐이다.- 게이트가 축을 대조한다(GATE-C8).
axes가 있으면checklist-gate.sh가score를 축에서 재계산해 불일치와 "test=0인데 임계 이상"을 차단하고,tests가 있으면 test 축을 실행 결과에서 다시 파생해 대조한다.axes없는 항목은 기존대로score만 본다(구형·타 에이전트 체크리스트 하위호환).
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.
- 1mo ago Changed · +5 lines af93f212f595
- 1mo ago Changed · +1 lines 08fc1d87cd9f
- 1mo ago First seen · 103 lines · 83 tokens per session scan A b08c48a550d4
checklist-loop is a skill published in the GitHub repository claude-code-expert/carve-harness (15 stars, last pushed 1mo ago), licensed MIT. It adds 83 tokens to every session and 2,444 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-08-30.
Other skills, from other repositories
research-engineer
An uncompromising Academic Research Engineer. Operates with absolute scientific rigor, objective criticism, and zero flair. Focuses on theoretical correctness, formal verification, and optimal implementation across any required technology.
tika-eval-compare
Compare extracts from two Tika builds over a corpus to detect regressions in content, encoding, exceptions, and embedded-document handling. Use for "compare before/after extracts", "eval this change against the corpus".
cua-driver
Use Cua Driver for desktop or browser tasks that are awkward or unavailable through Bash/APIs, or when the user explicitly wants GUI interaction: app testing, visual bug reproduction, form filling, calendar entry, screenshots, and demo recording. Also covers Cua setup; not OpenAI Codex Computer Use or web research.
jetson-validate-image
Use after jetson-flash-image to run static BSP checks, on-target smoke/regression tests on a flashed DUT, or both. Not for build or flash steps. Triggers: validate bsp, on-target validation.
atmos-validation
Validate Atmos projects, components, arbitrary JSON Schema inputs, EditorConfig, and GitHub Actions; use affected-file selection and native CI annotations.
skill-benchmark
Benchmark AI skill effectiveness by measuring implementation quality against legacy constraints.