Novel Writer logo

Novel Writer

Community
Yunshiro
novel-writer

Create, continue, revise, or audit Chinese fiction through a staged workflow covering requirements, character DNA, outlining, emotional beats, chapter drafting, deterministic manuscript checks, layered revision, and long-form continuity. Use when the user asks to write a novel or story, develop a commercial web-fiction project for platforms such as Fanqie, Qimao, or Qidian, create literary fiction or contest submissions, continue an existing manuscript, remove AI-like prose patterns, or inspect novel chapters for immersion and pacing.

Overview

PublisherYunshiro
Repositoryyunn-skills
Skill namenovel-writer
Stars
54
Forks
17
Bundled files
8
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 8 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by Yunshiro on GitHub. Read the source before you install it.

Installation

Install the Novel Writer AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/Yunshiro/yunn-skills/tree/main/skills/novel-writer
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/Yunshiro/yunn-skills.git /tmp/yunn-skills
mkdir -p .claude/skills
cp -r /tmp/yunn-skills/skills/novel-writer .claude/skills/novel-writer
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Novel Writer in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Novel Writer on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Novel Writer is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

小说创作工坊

将小说创作拆成可确认、可恢复、可验证的六个阶段。不要一次性跳过规划直接生成全稿,除非用户明确要求跳过确认。

核心优先级

按以下顺序解决冲突:

  1. 保证人物行为、故事因果和设定一致。
  2. 保证视角贴身和读者代入感。
  3. 移除模具句、机械排比与编排痕迹。
  4. 调整节奏、字数和格式。

商业网文中,代入感高于去 AI 味。不要为了减少心理活动而删掉主角的即时感知、身体反应、危险判断和明确欲望。

商业网文的爽感结构与代入感同属第一优先:大纲与节拍阶段对照 references/shuanwen-tech.md 的爽感自检,正文阶段落实打脸四拍、差异感与低调装逼。

资源路由

只读取当前阶段需要的参考文件:

当前任务必须读取
需求采集、角色 DNA、大纲准备、情绪节拍references/planning.md
商业网文的平台结构、黄金三章、爽点、冲突升级、代入感references/commercial-mode.md
商业网文的爽感引擎:差异感、迪化、异性倾慕、低调装逼、反转节奏、爽感自检references/shuanwen-tech.md
纯文学、征文、自留或文学模式references/literary-mode.md
正文创作、单章检查、全稿体检或分层润色references/writing-and-revision.md
达到 13 章或 5 万字、按卷创作、跨会话续写references/long-form-continuity.md

解析本 Skill 所在目录为 <skill-dir>。运行检查时调用 <skill-dir>/scripts/check_novel.py,不要把脚本重新写进提示词或项目目录。

模式判断

阶段一必须确认平台或用途,再决定模式。两套结构规则不可混用。

用户目标模式结构重点
番茄、七猫、得间、疯读等免费阅读平台商业网文黄金三章、密集反馈、锯齿上行情绪
起点、掌阅、QQ 阅读等付费订阅平台商业网文同样执行商业结构,允许略慢铺垫
纯文学、征文、文学刊物、个人自留文学人物选择、价值转折、潜台词与余韵

用户无法判断时,说明假设并默认商业网文模式。

篇幅判断:

  • 不超过 12 章且少于 5 万字:中短篇,逐章确认后创作。
  • 达到 13 章或 5 万字:长篇,按每卷 3 至 5 章规划;卷前确认一次,卷内连续创作。

输出目录

所有创作文件保存到当前项目的 output 目录。需要时创建目录:

text
output/
├── 章节正文/
│   ├── 第1章·章节标题.md
│   └── ...
└── 创作材料/
    ├── 书名-角色DNA档案.md
    ├── 书名-故事大纲.md
    ├── 书名-情绪节拍表.md
    ├── 书名-创作进度汇总.md    # 长篇
    └── ...

正文文件除章节标题外不使用 Markdown 语法。角色、大纲、节拍和进度文档可以使用 Markdown 表格与列表。

六阶段工作流

阶段一:需求采集

读取 references/planning.md。只询问用户尚未提供的信息,至少明确:

  • 平台或用途、类型、故事内核。
  • 商业网文的金手指机制、限制和代价。
  • 背景、视角、人物数量和参考作品。
  • 总字数、章节数和单章目标字数。

输出《故事规划确认表》,注明模式、篇幅模式、已知事实、默认假设和待确认项。用户确认后进入阶段二。

阶段二:角色 DNA

继续使用 references/planning.md。为主要人物建立能从过去经历推导到恐惧、渴望和行为的档案。

商业网文额外指定:

  • 反派的轻蔑表达习惯。
  • 配角的明确反馈功能。
  • 1 至 2 名具名情绪代理人。
  • 主角固定的身体感知清单和内心独白语气。

保存角色 DNA,向用户展示摘要并确认后进入阶段三。

阶段三:故事大纲

先读取 references/planning.md,再按模式读取:

  • 商业网文:references/commercial-mode.md + references/shuanwen-tech.md。
  • 文学模式:references/literary-mode.md。

商业网文不得使用“激励事件可以拖到全书四分之一处”的慢结构替代黄金三章。文学模式不得硬塞每 2000 字爽点或当面打脸。

保存故事大纲,完成对应模式的自检。用户确认后进入阶段四。

阶段四:情绪节拍

读取 references/planning.md 的节拍规则,商业网文再对照 references/shuanwen-tech.md 的爽感自检(快感节奏、反转铺垫、期待感钩子)。为每章写出事件目标、情绪变化、价值转折、代价、信息变化、伏笔和具体章末事件。

长篇先确认卷级节拍,再只展开当前卷。保存节拍表并等待确认。

阶段五:正文创作

读取 references/writing-and-revision.md,并保持对应模式参考文件在上下文中(商业网文含 references/shuanwen-tech.md,落实打脸四拍、差异感与低调装逼)。长篇还要读取 references/long-form-continuity.md 和当前进度文件。

每章按以下顺序完成:

  1. 读取角色 DNA、大纲和当前章节拍。
  2. 按目标字数反推场景拍子,一稿写入目标上下 200 字。
  3. 保存到 output/章节正文/第N章·[标题].md。
  4. 运行单章检查:
bash
python3 <skill-dir>/scripts/check_novel.py chapter \
  "./output/章节正文/第N章·标题.md" \
  --target 2500 \
  --hero "主角名"
  1. 修复脚本报告的问题并重新运行,直到机械指标通过。
  2. 人工通读视角、因果、人物利益、数字常识、对话声音和动作复用。
  3. 中短篇向用户报告本章摘要、字数、检查结果和下一章目标,等待确认。
  4. 长篇更新进度、人物状态、能力、伏笔和精确续写入口;卷内继续下一章。

将示例命令中的目标字数和主角名替换为当前项目值。脚本非零退出表示仍有机械指标未通过,不要忽略。

阶段六:分层润色

读取 references/writing-and-revision.md,严格按顺序执行:

  1. 代入感润色:补视角感知、即时情绪、身体感官和人味。
  2. 去 AI 味润色:删模具、重复解释、机械排比和编排痕迹。
  3. 节奏压缩:删除无功能段落和不改变局势的对话。
  4. 对话优化。
  5. 环境与心理活动优化。

商业网文必须执行前三层;其余层根据章节重点执行。文学模式也执行去 AI 味检查,但不强制商业爽点与章末悬念。

全稿完成后运行:

bash
python3 <skill-dir>/scripts/check_novel.py manuscript \
  "./output/章节正文" \
  --hero "主角名"

所有章节目标相同时添加 --target 2500。脚本通过后仍执行参考文件中的人工终检。

长篇续写

长篇模式读取 references/long-form-continuity.md。每章立即更新进度,不等卷末凭记忆补写。

新会话恢复时依次读取快速记忆、进度汇总、当前卷节拍、相关角色 DNA 和上一章全文。先向用户报告上一章停点、当前状态和下一章目标,再继续写作。不要重新规划已经确认的内容。

完成标准

交付前确认:

  • 所有必要确认节点均已通过,或用户明确要求跳过。
  • 角色、大纲、节拍和正文文件路径符合规范。
  • 每章机械检查通过,且完成无法脚本化的人工通读。
  • 商业网文已通过 references/shuanwen-tech.md 的爽感自检。
  • 伏笔、人物状态、能力限制、时间、距离和数量前后一致。
  • 长篇的进度与续写入口已经更新到最后完成的章节。
  • 最终报告列出已完成文件、总字数、检查结果和仍需用户决定的问题。

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Novel Writer AI skill do?

Create, continue, revise, or audit Chinese fiction through a staged workflow covering requirements, character DNA, outlining, emotional beats, chapter drafting, deterministic manuscript checks, layered revision, and long-form continuity. Use when the user asks to write a novel or story, develop a commercial web-fiction project for platforms such as Fanqie, Qimao, or Qidian, create literary fiction or contest submissions, continue an existing manuscript, remove AI-like prose patterns, or inspect novel chapters for immersion and pacing.

Why use Novel Writer on TypingMind?

Because you install it once and use it with any model. Novel Writer is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Novel Writer in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/Yunshiro/yunn-skills/tree/main/skills/novel-writer. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Novel Writer?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Novel Writer?

As many as you like. As long as a model supports skills, you can use Novel Writer with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Novel Writer AI skill free?

Yes. It is published on GitHub by Yunshiro under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇