agentmods.dev

review-doc

review-doc is a skill for Claude Code from kohj1018/agentic-dev-harness. It costs 31 tokens per session (1,444 once invoked), scanned A, original, MIT.

A document-review process for finding missing requirements, contradictions, vague statements, hidden complexity, and overlooked edge cases.

In plain words
What is it for?
Use it to review project documents, report serious mismatches, check length limits for AGENTS.md, and compare selected decision tables with their source records.
Why use it?
It makes documents easier to understand and checks whether their contents match their stated mode or purpose.

Skill for Claude Code

Written for Claude Code: allowed-tools in frontmatter. Also seen: agent in frontmatter; reads .claude/ paths; mentions AGENTS.md.

not rated 2repo 10d ago A scan 31 tokens original MIT

Good fit Use it to review project documents, report serious mismatches, check length limits for AGENTS.md, and compare selected decision tables with their source records.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/kohj1018/agentic-dev-harness/review-doc
View source ↗ kohj1018/agentic-dev-harness
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.

Any agent
npx skills add kohj1018/agentic-dev-harness --skill review-doc
Clone the repo
git clone --depth 1 https://github.com/kohj1018/agentic-dev-harness

Made for: Claude Code.

Wrote 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.

agentmods badge for review-doc

README.md
[![agentmods](https://agentmods.dev/badge/skills/kohj1018/agentic-dev-harness/review-doc.svg)](https://agentmods.dev/skills/kohj1018/agentic-dev-harness/review-doc)
Your own site
<a href="https://agentmods.dev/skills/kohj1018/agentic-dev-harness/review-doc"><img src="https://agentmods.dev/badge/skills/kohj1018/agentic-dev-harness/review-doc.svg" alt="Measured on agentmods" height="20"></a>
Per session 31 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,444 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00031 $0.01444
Opus 5 $0.00015 $0.00722
Sonnet 5 $0.00006 $0.00289
Haiku 4.5 $0.00003 $0.00144

Measured 7d ago against content hash 20178ee65e2c, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

review-doc 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 7d 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.

.claude/skills/review-doc/SKILL.md · 63 lines

What it actually says

너의 역할은 입력 경로의 문서를 비판적으로 검토하는 것이다.

Codex: 본 skill은 wrapper 미보유(자연어 호출) — Codex에서는 "Follow .claude/skills/review-doc/SKILL.md"로 호출한다(목록 SSOT = README, ADR-010#amend-3·#amend-4).

입력:

  • $ARGUMENTS에는 검토할 문서의 경로가 들어온다(예: docs/10-charter/PROJECT_CHARTER.md).

반드시 먼저 할 일:

  1. 입력 경로의 문서를 읽는다.
  2. 필요하면 관련 상위/하위 문서를 함께 읽어 맥락을 파악한다.

검토 항목:

  • 누락된 요구사항
  • 모순된 진술
  • 모호하거나 검증 불가능한 표현
  • 숨은 복잡도
  • 빠진 엣지 케이스
  • 본문 내용이 첫 줄의 모드 라벨(> 모드: ...)과 정합한지 점검.
  • mismatch 발견 시 docs/40-validation/IMPROVEMENT_GUIDE.md에 P1 severity로 보고.

아래 항목은 검토 대상 문서가 그 항목의 대상일 때만 적용한다 (review-doc 은 단일 문서 도구 — 예: AGENTS.md 검토 시 길이 점검 / boilerplate README 검토 시 표 점검 / docs/00-meta/ 문서 검토 시 파일 수 점검. repo 전체 상시 감사는 /stabilize-milestone deterministic preflight 책임 — stabilize-milestone SKILL "review-doc 책임 분담" 단락 정합):

  • AGENTS.md 길이 점검: 100줄 초과 시 IMPROVEMENT_GUIDE에 P0 severity로 보고. 80~100줄 사이는 P1.
  • docs/90-decisions/boilerplate/README.md의 Reserved / Parked / Dropped 번호 표가 git log의 실제 누락 번호와 일치하는지 점검. 새 dropped 번호 발견 시 P2 보고.
  • docs/90-decisions/boilerplate/README.md ADR 표의 Amendments 컬럼이 각 ADR 본문의 실제 ## Amendment N 단락 — 코드펜스(```)·<!-- --> 주석 밖의 것만(예시·주석 헤딩 제외) — 과 일치하는지 점검. 누락 발견 시 P1 보고.
  • docs/00-meta/ 파일 수가 ADR-012의 6개 원칙과 일치하는지 (_templates/는 카운트 제외). 위반 시 P0 보고.
  • 검토 대상이 ## Surfaces 블록을 가진 ADR이면, 등재된 각 surface 파일에 해당 ADR-NNN 역참조가 있는지 spot-check. 누락 시 P1 [Surface-backref] 보고 (ADR-045#d4).
  • reviewer 위임 시 입력에 review surface: doc 를 명시한다 (reviewer.md 의 Document Consistency 차원 정합 — Clean Code / Scope Discipline / Document Consistency 3 차원 중 doc surface 선택).

Write 범위 제한 (수정 대상 파일 제한 — frontmatter allowed-tools 와 직교):

  • frontmatter allowed-tools 의 Write/Edit 는 도구 호출 가능성 만 정한다 (그래야 IMPROVEMENT_GUIDE 에 기록 가능).
  • 본문은 수정 대상 파일 을 docs/40-validation/IMPROVEMENT_GUIDE.md 단일 파일 로 제한한다 — 그 외 어떤 파일도 Write/Edit 금지.
  • 본문 외 변경이 필요해 보이면 출력에 "후속 task 권장" 텍스트만 남긴다 — workitem finding은 구현 전 task·매핑·의존성 결함이면 /validate-plan M<N>→/repair-plan M<N>, M/F/prototype 계약 결함이면 사용자 보고, 구현 시작 뒤 기존 약속 결함이면 소유 task repair, 담당 task 없음·새 범위면 사용자 보고+다음 M 후보로 분기한다(ADR-057#amend-3 결정 6). review-doc은 계속 report-only다.
  • 위반 발견 시 IMPROVEMENT_GUIDE.md 에 self-report (예: P1 [Self-violation] review-doc edited <file>) + 다음 라운드 stabilize 가 회수.

마지막 출력:

  • 결과를 P0, P1, P2로 나눈다.
  • 어떤 섹션을 어떻게 수정하면 좋을지 구체적으로 제안한다.
  • 상위 설계 문제와 하위 구현 문제를 구분한다.
  • 막연한 칭찬은 하지 않는다.
  • 시간/턴이 부족하면 확인된 범위까지의 핵심 판단만 요약하고 종료한다.
  • 다음 단계 (WORKFLOW.md "스킬 종료 시 다음 단계 출력 contract" 양식 정합):
    • 기본 권장: P0 finding 이 0건이면 후속 skill 없이 종료. P0/P1 이 있으면 검토 대상 문서 종류별 분기.
    • 분기 옵션 (해당 시 ≤3):
      • workitem 문서 (milestone/feature/task) 면: 구현 전이면 /validate-plan M<N>→/repair-plan M<N>, 계약(M/F/prototype) 결함이면 사용자 보고, 구현 시작 뒤면 소유 task repair 또는 사용자 보고+다음 M 후보
      • charter / architecture / ADR 이면: 메인이 architect 단발 위임으로 갱신
      • AGENTS.md / 운영 문서이면: 사용자 직접 수정 (Living Doc 갱신)
    • 프롬프트 동봉 권장:
      • 본 review 출력의 P0/P1 finding 라벨 + 라인 위치 (수정자의 컨텍스트 회수용)
      • 본 review 가 건너뛴 영역 (시간/턴 부족 시) — 다음 라운드 review-doc 호출의 우선순위 입력

Context 정책 (ADR-019)

반드시 먼저 읽을 파일은 최소 충분. 추가 ADR/architecture 섹션은 task 본문에서 발화 시 인용 — 사전 fork-load 금지.

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. 7d ago First seen · 63 lines · 31 tokens per session scan A 20178ee65e2c

Subscribe to this mod's changes

review-doc is a skill published in the GitHub repository kohj1018/agentic-dev-harness (2 stars, last pushed 10d ago), licensed MIT. It adds 31 tokens to every session and 1,444 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.

Claude CodeWriting and docsagentic-workflowai-agentsboilerplatecodex-clideveloper-tools
Related

Other skills, from other repositories

sc-skill

Capture deterministic macOS screenshots for testing, docs, release notes, and marketing assets. Use when asked to automate app screenshots, batch-generate screenshot sets, standardize window sizing/composition, or choose between Peekaboo and native macOS screenshot tooling.

jazzyalex/agent-sessions · 55 tokens

handover

Use when wrapping up or capturing the current state of a coding session — writes a short, dated entry to the repo's RepoHandover.md so a future agent or you can resume without grepping archived sessions. Triggers on "handover", "hand off", "write handover", "capture state", "checkpoint this session".

jazzyalex/agent-sessions · 71 tokens

agent-session-format-check

Verify agent session format compatibility for Agent Sessions. Use when any agent CLI updates, when monitoring flags drift, or when bumping max verified versions (fixtures + docs + tests). Covers session schema, usage/limits tracking, storage backends, and discovery path contracts for all supported agents.

jazzyalex/agent-sessions · 62 tokens

agent-support-matrix

Maintain Agent Sessions agent support matrix and JSON/JSONL parsing compatibility. Use when checking upstream agent releases for session format changes, updating max verified versions in docs/agent-support/agent-support-matrix.yml, or updating docs/agent-json-tracking.md and fixtures/tests.

jazzyalex/agent-sessions · 59 tokens

do-it-architecture

Use when authority, ownership, contracts, dependency boundaries, migration, recovery, or structural deletion can change a design.

tdwhere123/do-it · 28 tokens

do-it-code-quality

Use when designing, changing, or debugging code to locate causal ownership and close the affected behavior.

tdwhere123/do-it · 24 tokens