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.
/plugin marketplace add chaserr/novel-craft/plugin install novel-craftWrote 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/chaserr/novel-craft/novel-write)<a href="https://agentmods.dev/skills/chaserr/novel-craft/novel-write"><img src="https://agentmods.dev/badge/skills/chaserr/novel-craft/novel-write/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/chaserr/novel-craft/novel-write"><img src="https://agentmods.dev/badge/skills/chaserr/novel-craft/novel-write.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 | $0.00073 | $0.03540 |
| Opus 5 | $0.00036 | $0.01770 |
| Sonnet 5 | $0.00015 | $0.00708 |
| Haiku 4.5 | $0.00007 | $0.00354 |
Grade A, and why
novel-write 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 today.
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 — 189 lines — stays where its author put it; the contents beside it link to each section on GitHub.
novel-write
用途：长篇小说项目里"写一章"的标准工作流。在没有 RTK.md 的项目里禁止执行——必须先跑 novel-init。
输入要求
用户应说明：
- 章节号（第 X 章）。如未说，从
章节大纲.md推算下一未写章节。 - 章节名（可选）。如未说，从
章节大纲.md取。 - 本章特殊要求（可选）：必须出现的事件 / 必须埋的伏笔 / 必须收回的伏笔 / 不能出现的内容。
执行流程
第 1 步：上下文加载（必须，按顺序）
全部读取，不可跳过。如缺失任一文件，先提示用户走 novel-init 或手动补齐：
RTK.md— 项目规则、题材气质、文风底线（最高优先级）AGENTS.md— 项目入口（如指向其他规则文件，一并读取）小说大纲.md— 确认本章在全书的位置和功能章节大纲.md— 本章的节点和情绪目标前情梳理.md— 必须完整读完最近 3 章的摘要，不可只看本章前一章- 本章直接相邻的章节正文：至少回看上一章原文，必要时回看最近 2-3 章原文
伏笔清单.md— 当前活跃伏笔、需要本章兑现的伏笔- 本章涉及人物的
人物档案/*.md— 说话方式、当前情感状态、关系进度 经典语录.md— 已有金句，避免重复相似句式- 写作技法（两层，优先级：项目 > 通用）
- 项目专属：
写作技巧/*.md（项目自己沉淀的题材专属技法）— 优先 - plugin 通用：
${CLAUDE_PLUGIN_ROOT}/skills/novel-write/reference/*.md— 兜底 - 通用 reference 包含：01 运镜 / 02 感官 / 03 细节 / 04 心理 / 05 场景四字诀 / 06 信息密度 / 07 章末结构 / 08 画面感总则 / 09 期待值管理 / 10 情绪控制力
- 按需读取：不每次全读，按本章场景类型选 1-2 篇查阅（详见各 reference 末尾的"题材适配建议"）
- 先判适配再用：读某篇前先看它顶部的「适用边界」块 +
reference/README.md的「技法×网文适配 速查表」，技法不匹配本书题材/节奏就不用——网文默认别把 01/02/04/08 当每段必用
- 项目专属：
审稿报告/*.md（如存在）— 历轮审稿提炼的"作者习惯性弱点"，写作时主动规避
第 2 步：本章写作目标确认（不可跳过）
读完上下文后，用一段结构化输出向用户确认：
【第X章 写作确认】
本章在全书位置：第N幕 / 第M阶段 / 占总章节 X%
情绪基调：↑/↓/→/⚡
主要节点：
1. ...
2. ...
需要兑现的伏笔：
- 伏笔A（埋于第Y章）
需要新埋的伏笔：
- 待埋（轻、自然）
人物状态承接：
- 主角：从第X章 [情绪/位置] 接续
- 配角：...
时代/环境细节（如题材需要）：
- ...
本章字数预期：~XXXX 字
是否开始生成初稿？（如有调整请告诉我）
用户确认后才进入第 3 步。
第 3 步：生成初稿
按以下原则写：
题材气质遵守
- 严格执行
RTK.md里的"文风要求"和"感情线/主线要求" - 用户提供的"核心气质 ≤5 个"是红线，不可违背
- 如题材有专属符号库（在
写作技巧/下），优先使用
通用写作铁律
| 编号 | 铁律 | 反例 | 正例方向 |
|---|---|---|---|
| W1 | 一个情绪瞬间只写一次，不要写完场景再用一句"她忽然意识到……"做内心独白复述 | "她忽然明白，有些重逢……" | 一个动作 + 一个余响 |
| W2 | 章节结尾不连续使用"感悟收束"。可用"未完动作 / 物件落定 / 群消息切断 / 场景叠加"等替代 | "她忽然觉得，所谓青春……" | "抽屉合上时，发出一声很轻的'咔'。" |
| W3 | 高频套话黑名单（具体清单写在 RTK.md）每章上限 1 次，最好 0 次 | "心跳漏了一拍" / "岁月静好" | 用具体感知细节代替 |
| W4 | 对话必须符合角色身份和年龄。遮住名字看不出是谁说的 = 失败 | 所有角色都说一样的话 | 词汇量 / 句长 / 口头禅 / 思维方式分化 |
| W5 | 场景描写至少完成一项功能：情绪 / 性格 / 推进 / 主题。否则可删 | 纯背景板 | 环境作为情绪容器 |
| W6 | 每段至少有一个信息量推进。不写空转段落 | 反复抒情 | 推进事件、关系、内心或细节 |
| W7 | 视角自洽。第三人称限定/第一人称里，主角不感知的事不写 | 主角看不见但描写另一房间 | 留空 or 切视角章节 |
| W8 | 主角初遇超常/危机要走真实心理弧线：不信→目睹代价→惊慌→慢慢适应。第一次别像老手，别一上来就冷静解读规则/破局；首次多半靠碰巧活，别人的死/代价才教会他 | 普通人第一次进诡异副本就淡定推理、精准避开 | 先怕、先蒙、先付代价，能力和冷静是后面挣出来的 |
| W9 | 禁用职业/专业术语给场景做旁白——氛围杀手，一秒出戏。角色的专业只藏在思维方式里（对矛盾敏感、会做假设再验证），不在词汇里 | "他脑子里冒出：边界溢出、数组越界" | 用大白话写出同一个"不对劲"，专业感靠推理逻辑体现 |
| W10 | 金手指/能力升级要有**"看得见的质变时刻"：对照（"这个我以前做不到"）＋旁人惊愕＋打法变了；且必带代价**，无代价=无敌=张力崩。软成长尤其要补这三样 | 主角悄无声息变强，读者无感 | 一场戏专门让他做到过去做不到的事，并为此付出可见代价 |
| W11 | 副本/单元流防"换地图"：副本只是舞台，真正在爬的是多条线——金手指线 / 真相线 / 格局线 / 敌人线 / 羁绊线 / 主角身份转变线，每过一个副本各升一格 | 一个场景换一个场景，主角原地踏步 | 换考场，不换成长；读者追的是六条线一起往上走 |
| W12 | 事实记账（配合 novel-sync）：本章产生的"既成事实"——谁死了、揭了哪条规则/设定、埋/收了哪个伏笔、人物状态与代价变化、时间推进——必须回填，不许和已写矛盾。文笔可自由发挥，事实不可 | 上一章死的人下一章复活、已揭的规则装作没揭 | 写完立即 novel-sync；把"已揭规则账/人物状态账/伏笔账"记进前情梳理 |
What ships with it
11 files 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.
- reference/01-运镜.md 4.1 KB
- reference/02-感官.md 4.3 KB
- reference/03-细节.md 3.6 KB
- reference/04-心理.md 4.3 KB
- reference/05-场景四字诀.md 6.7 KB
- reference/06-信息密度.md 3.6 KB
- reference/07-章末结构.md 4.9 KB
- reference/08-画面感总则.md 3.5 KB
- reference/09-期待值管理.md 4.5 KB
- reference/10-情绪控制力.md 6.7 KB
- reference/README.md 5.2 KB
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.
- today Changed · +1 lines 0c84cc22a465
- yesterday Changed · +20 lines b16d4d57348f
- 9d ago First seen · 168 lines · 73 tokens per session scan A c730e8f11efa
novel-write is a skill published in the GitHub repository chaserr/novel-craft (2 stars, last pushed yesterday), licensed MIT. It adds 73 tokens to every session and 3,540 once invoked, about $0.0004 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
story-import
A tool for turning an existing novel into a structured writing project. It analyzes the book and organizes its characters, settings, plot plans, chapters, and tracking data for later writing.
story-review
A review workflow for finding story problems from several viewpoints, including issues with structure, characters, wording, and fictional world rules. It can use multiple reviewer agents or work alone.
story-deslop
A writing editor for Chinese web novels that detects writing patterns often associated with AI-generated text and makes the prose feel more natural.
story-long-analyze
A long-form fiction analysis workflow for breaking down a novel’s opening chapters, characters, pacing, turning points, relationships, and overall structure.
story-long-write
A Chinese-language coaching workflow for creating long online novels, from the initial idea and outline through characters, plot threads, and chapter drafts. It also supports continuing, revising, or rewriting specified chapters.
story-short-write
A workflow for creating short online fiction, from story ideas and outlines through a complete draft, with emphasis on emotional tension and pacing.