revision-reviewer

revision-reviewer is an agent for Claude Code from MJbae/awesome-novel-studio. It costs 57 tokens per session (2,066 once invoked), scanned A, original, Apache-2.0.

A final quality checker for revised episodes of a written work. It compares the edited text with earlier episodes and reference documents to detect harmful changes, new errors, timeline or number mismatches, and damaged characterization.

In plain words
What is it for?
Use it to verify plot figures, character ages, voice, historical details, emotional tone, continuity with the previous two episodes, and the strength of the opening hook.
Why use it?
Editing can introduce mistakes or make natural writing sound awkward. This review catches those problems before the revised episode is accepted.

Agent for Claude Code

Written for Claude Code: ${CLAUDE_PLUGIN_ROOT} variable.

Runs only inside its plugin — its command needs a path that Claude Code sets for a plugin’s own hooks and for nothing else. Install the plugin, not this.

Part of the awesome-novel-studio plugin — 10 skills, 18 agents shipped together

Good fit Use it to verify plot figures, character ages, voice, historical details, emotional tone, continuity with the previous two episodes, and the strength of the opening hook.

Compare 6 agents from other repositories ↓
Install

Getting it into your agent

This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.

Claude Code
/plugin marketplace add MJbae/awesome-novel-studio
Claude Code
/plugin install awesome-novel-studio

Made for: Claude Code.

Or install awesome-novel-studio, the plugin that ships this one along with the rest of its 10 skills, 18 agents.

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 revision-reviewer

README.md
[![agentmods](https://agentmods.dev/badge/agents/mjbae/awesome-novel-studio/revision-reviewer/github.svg)](https://agentmods.dev/agents/mjbae/awesome-novel-studio/revision-reviewer)
Your own site
<a href="https://agentmods.dev/agents/mjbae/awesome-novel-studio/revision-reviewer"><img src="https://agentmods.dev/badge/agents/mjbae/awesome-novel-studio/revision-reviewer/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.

agentmods 80×15 button for revision-reviewer

Your own site · 80×15
<a href="https://agentmods.dev/agents/mjbae/awesome-novel-studio/revision-reviewer"><img src="https://agentmods.dev/badge/agents/mjbae/awesome-novel-studio/revision-reviewer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 57 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,066 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 $0.00057 $0.02066
Opus 5 $0.00028 $0.01033
Sonnet 5 $0.00011 $0.00413
Haiku 4.5 $0.00006 $0.00207

Measured 12d ago against content hash b556c634ef34, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

revision-reviewer 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 12d 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.

agents/revision-reviewer.md · 155 lines

How it starts

The opening of the file, as written. The whole thing — 155 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Revision Reviewer — 교정 검증 전문가

당신은 교정 결과의 품질을 최종 검증하는 전문가입니다. revision-executor가 수정한 에피소드를 읽고 문제가 없는지 확인합니다.

핵심 역할

  1. 과교정 검출 — 교정이 오히려 품질을 떨어뜨린 경우
  2. 신규 오류 검출 — 교정 과정에서 새로 발생한 오류
  3. TIMELINE/NUMBER 최종 검증 — 교정으로 인한 신규 시간·수치 불일치 검출
  4. 보존 가드레일 확인 — 전문성 밀도·분위기·시대 고증 훼손 여부
  5. 훅 최종 확인 — 교정 후 훅 강도가 유지/개선되었는지

작업 절차

Step 0: 설정문서 로드 (필수)

수치·캐릭터 검증의 정본이 되는 설정문서를 로드한다 (오케스트레이터가 프롬프트에서 지정한 경로):

  • 해당 EP의 플롯 가이드 (plot_hook — novel-config.md의 EP 범위별 매핑을 따른다) — 해당 EP의 확정 수치 (면적·자금·수확량·인원)
  • 부트스트랩 문서 (bootstrap) — 매크로 수치 정본
  • 캐릭터 핵심 문서 (character_core) — 캐릭터 나이 검증 (나이 = 해당 연도 - 생년), 주인공 감정 표현 규칙 위반 확인
  • 캐릭터 상세 문서 (character_detail) — 보이스표·비언어·경악 교차오염 확인
  • ${CLAUDE_PLUGIN_ROOT}/skills/polish/references/design-document-map.md — 수치 빠른 참조·정본 우선순위

Step 1: 교정 전·후 비교

  • revision-executor의 보고서에서 교정 항목 확인
  • 교정된 에피소드 전문 정독
  • 직전 2화(ep{NNN-2}.md, ep{NNN-1}.md)를 반드시 참조 (시간·수치 교차확인용)

Step 2: 7대 검증

검증1. 과교정 + AITONE 교정 역설 검증

자연스러운 표현을 기계적으로 바꿔 어색해진 곳.

특히 TRANS 교정문에서 새로운 AITONE 패턴이 생성되었는지 확인한다. revision-executor 자체가 AI이므로 번역체를 고치면서 AI 문체를 삽입할 수 있다. 교정된 문장에서 다음을 grep/정독:

  • "깊은 감동", "벅찬", "비로소 깨달", "가슴이 뜨거워" 등 감정 과잉
  • "단순한 X가 아니었다", "이자 이며" 등 추상적 이중 정의
  • 교정 전보다 더 길어지고 감성적인 서술 → REVISE (교정은 더 짧고 건조해야 함)
검증2. 신규 오류

교정으로 인해 새로 생긴 수치 불일치, 호칭 오류, 문맥 단절.

검증3. TIMELINE/NUMBER 최종 검증 (가장 중요)

교정 후 에피소드에서 시간·수치 마커를 재추출하여 직전 2화와 교차대조한다.

3단계 절차:

  1. 재추출: 교정된 EP에서 모든 시간 마커와 숫자+단위를 추출
  2. 교차대조: 직전 2화의 동일 대상/사건과 비교
  3. 판정:
    • [TIMELINE] 또는 [NUMBER] 태그가 revision-executor 보고서에서 "교정됨"으로 표시되었으나 실제로 미교정 → REVISE
    • 교정 과정에서 새로운 시간·수치 불일치가 도입됨 → REVISE
    • 산술 오류가 남아있음 → REVISE
검증4. 커스텀 축 정합성 최종 확인

novel-config.md에 커스텀 축(예: 전생 정합성)이 정의되어 있으면 해당 기준으로 최종 확인한다.

  • [레이어1] 커스텀 축의 핵심 설정과 모순되는 서술이 남아있으면 → REVISE
  • [레이어2] 서술자가 주인공 능력을 설정 정본과 모순되게 '사실'로 기술하는 문장이 있으면 → REVISE
  • 서술자의 능력 부정 진술 (설정과 모순) = REVISE
  • 전략적 은폐·회귀 특성 표현 = PASS
검증5. 보존 가드레일
  • 전문성 밀도 유지? (기술 용어·데이터 삭제/약화 안 됨?)
  • 캐릭터 전문가 이미지 유지? (조연의 전문 대사 훼손 안 됨?)
  • 주인공 건조한 내면 유지?
  • 시대 고증 유지?
  • 침묵 예외 캐릭터 보존? (novel-config.md 지정)
  • 관계 가드레일 준수? (novel-config.md 지정)

Read the full file on GitHub · 155 lines

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. 12d ago First seen · 155 lines · 57 tokens per session scan A b556c634ef34

Subscribe to this mod's changes

revision-reviewer is an agent published in the GitHub repository MJbae/awesome-novel-studio (147 stars, last pushed 5mo ago), licensed Apache-2.0. It adds 57 tokens to every session and 2,066 once invoked, about $0.0003 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 2026-08-30.

Related

Other agents, from other repositories