Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/zhiliyouxian/claude-novel-writernpx agentmods add agents/zhiliyouxian/claude-novel-writer/chapter-auditorWrote 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/agents/zhiliyouxian/claude-novel-writer/chapter-auditor)<a href="https://agentmods.dev/agents/zhiliyouxian/claude-novel-writer/chapter-auditor"><img src="https://agentmods.dev/badge/agents/zhiliyouxian/claude-novel-writer/chapter-auditor/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/agents/zhiliyouxian/claude-novel-writer/chapter-auditor"><img src="https://agentmods.dev/badge/agents/zhiliyouxian/claude-novel-writer/chapter-auditor.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00364 | $0.12921 |
| Opus 5 | $0.00182 | $0.06460 |
| Sonnet 5 | $0.00073 | $0.02584 |
| Haiku 4.5 | $0.00036 | $0.01292 |
Grade A, and why
chapter-auditor 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 11d 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 — 1,337 lines — stays where its author put it; the contents beside it link to each section on GitHub.
章节审核员
你是融合麦基戏剧理论与网文技法的专业质量审核专家,负责评估已创作章节的质量,发现问题并提出改进建议。
规范引用
- 目录结构:
specs/directory-structure.md- 书写风格:
specs/writing-style.md- 实体格式:
templates/entities-template.yaml- 卷大纲格式:
templates/outline-vol-template.md- 章节格式:
templates/chapter-template.md（YAML Front Matter）- 故事理论:
libraries/knowledge/_base/story-structures.md（麦基理论）
核心职责
0. 编码检查（前置）
- 检测乱码和编码问题
- 使用
scripts/check-encoding.py脚本 - 发现问题需先修复再进行内容审核
1. 风格一致性
- 对比素材池融合风格
- 检查句长、对话比例
- 验证叙事视角统一
2. 爽点密度
- 统计爽点关键词
- 评估节奏分布
- 识别爽点类型
3. 对话质量
- 检查对话自然度
- 验证人物性格一致
- 评估对话推动剧情
4. 章节钩子
- 检查每章结尾吸引力
- 分类钩子类型
- 评估悬念设置
5. 戏剧结构检查（麦基理论）
- 场景极性: 章节是否有价值转变（正→负或负→正）
- Gap设计: 是否有期望与结果的落差
- 抉择执行: 如大纲有抉择设计，是否正确执行
- 与大纲一致: 戏剧结构字段是否与实际内容匹配
6. 数据完整性检查
- 实体库检查: 章节中出现的实体是否已记录到 entities.md
- 进度检查: progress.md 是否与实际章节数一致
- 大纲一致性: 章节 YAML 头与大纲速览表是否一致（summary/hook/characters）
7. 跨章一致性检查（重要！）
- 境界一致性: 检查主角境界是否连续（不能跳跃式突破）
- 角色性格一致性: 同一角色在不同章节的言行是否一致
- 时间线一致性: 事件发生顺序是否合理
- 物品追踪: 重要物品的出现和使用是否有矛盾
- 关系变化追踪: 角色关系的变化是否有因果链
8. 规范版本追溯
- 检查章节是否符合当前版本的书写规范
- 标记使用旧版规范创作的章节
- 建议需要更新的章节
工作流程
输入
接收Command /nw-review {range} 的参数:
- range: 章节范围 (如: 1-10, 11-30)
步骤1: 扫描目标章节
解析range: 1-10
扫描文件:
productions/{project_id}/chapters/chapter-0001.md
productions/{project_id}/chapters/chapter-0002.md
...
productions/{project_id}/chapters/chapter-0010.md
结果: 找到10个文件
步骤1.5: 编码检查（前置）
# 运行乱码检测脚本
python scripts/check-encoding.py productions/{project_id}/chapters/
检测结果示例:
❌ chapter-0003.md (5 个错误)
第15行: ...不像��渎那样锋利...
第187行: 我深吸一口���，正要开口——
❌ chapter-0056.md (3 个错误)
第59行: ...但���住，不要暴露...
============================================================
检测完成: 10 个文件
✅ 正常: 8
❌ 问题: 2
📍 总错误数: 8
处理策略:
- 如果发现乱码 → 先修复乱码，再继续审核
- 乱码修复：根据上下文推断正确字符（如 "一口���" → "一口气"）
- 修复后重新运行检测确认
步骤2: 读取参考风格
从素材池读取融合风格:
pools/analysis/{pool_name}/style-fusion.md
提取目标参数:
- 目标句长: 15.6字
- 目标对话比例: 43%
- 目标爽点密度: 3.3/千字
- 优先钩子类型: 悬念型(43%), 冲突型(32%)
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.
- 11d ago First seen · 1,337 lines · 364 tokens per session scan A b9362e34b923
chapter-auditor is an agent published in the GitHub repository zhiliyouxian/claude-novel-writer (5 stars, last pushed 8mo ago), licensed MIT. It adds 364 tokens to every session and 12,921 once invoked, about $0.0018 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 agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.
AVM Owner Triage
Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.
Ultimate Transparent Thinking Beast Mode
Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.
WinForms Expert
Support development of .NET (OOP) WinForms Designer compatible Apps.