fanqie-novel-skill
番茄小说专用长篇写作系统。融合番茄平台规则 + AI去味专项检查 + 商业审查。适用于在番茄小说平台进行长篇网文创作,支持120万字以上长线运营。
Safety scan
A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.
Badge for your README
Markdown
[](/listings/304769384-png--fanqie-novel-skill)HTML
<a href="/listings/304769384-png--fanqie-novel-skill"><img src="/api/listings/304769384-png--fanqie-novel-skill/badge.svg" alt="findsafeskills safety rating" /></a>Install
Install instructions from the README
As written by the repo's author. findsafeskills has not verified these steps.
Prerequisites
- AI coding assistant with skill support (e.g., Claude Code, Cursor) - Fanqie (Tomato) Novel Platform account
Installation
1. Download or clone this repository 2. Place the skill folder in your assistant's skill directory 3. Initialize a new novel project using the provided templates
Project Structure
After initialization, your novel project contains:
your-novel/
├── story_bible.md # World-building, settings, power system
├── book_rules.md # Platform rules + protagonist lock + taboos
├── outline.md # Volume outlines, chapter goals, payoff plans
├── current_state.md # Current world state (most important!)
├── pending_hooks.md # Foreshadowing & suspense tracking
├── handoff_current.md # Session handoff file (living document)
├── progress_tracker.md # Progress + rhythm quality dashboard
├── pattern_detection.md # Narrative pattern tracking
├── chapter_summaries.md # Per-chapter summaries
├── character_matrix.md # Character relationships & knowledge boundaries
├── style_guide.md # Style guide (with AI de-flavoring rules)
└── chapters/
└── ch001.mdPer-Chapter Workflow
BEFORE writing (5 mandatory): ├─ Read current_state.md → Confirm location/time/status ├─ Read outline.md → Confirm chapter goal, word count, is-climax ├─ Read pending_hooks.md → Check unresolved foreshadowing ├─ Check battle-scene-guide.md → If combat chapter, load checklist └─ Check dialogue-monitor.md → Confirm dialogue ratio trend AFTER writing (8 mandatory): ├─ 1. Word count check (2000-2500) ├─ 2. AI de-flavoring (smart replace) ├─ 3. Causal chain check ├─ 4. Behavior-rule consistency ├─ 5. Timeline continuity ├─ 6. Battle scene check (combat chapters) ├─ 7. Dialogue ratio check (≤40%) └─ 8. Talk-therapy detection (bosses can't be talked down) STATE UPDATE (mandatory): ├─ current_state.md ├─ pending_hooks.md ├─ handoff_current.md ├─ progress_tracker.md └─ pattern_detection.md
---
What was scanned
The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.
SKILL.md · Manifest · 348 lines
View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)
- ---
- name: fanqie-novel-skill
- description: 番茄小说专用长篇写作系统。融合番茄平台规则 + AI去味专项检查 + 商业审查。适用于在番茄小说平台进行长篇网文创作,支持120万字以上长线运营。
- ---
- # 番茄小说写作 Skill(Fanqie Novel Skill)
- 版本:**2.0.1** | 作者:**Jayce**
- **专为番茄小说平台打造的长篇网文写作系统。**
- ---
- ## v2.0.1 更新说明
- 本版本基于大量长篇网文实战创作与深度质量审阅的经验总结,进行重大架构升级。
- ### 核心问题回顾
- 长篇网文创作中反复出现的典型质量问题:
- | 问题 | 表现 | 严重程度 |
- |------|------|---------|
- | "话疗"泛滥 | 大量章节含说服类关键词,多个BOSS被对话解决 | 致命 |
- | 战斗描写崩溃 | 后期战斗场景几乎消失,意象重复率极高 | 致命 |
- | 对话占比失控 | 对话占比随篇幅推进持续飙升 | 严重 |
- | 节奏控制失败 | 篇幅超出故事承载力,后期注水严重 | 严重 |
- | 循环模式重复 | "觉醒-升级-新敌人"等固定模式无变化重复 | 中等 |
- | 反派降智 | 终极BOSS逻辑崩塌,轻易被说服 | 致命 |
- ### v2.0.1 核心改进
- | 改进项 | v1.1.1 | v2.0.1 |
- |--------|--------|--------|
- | 战斗场景 | 无专项检查 | 新增维度30"战斗场景审计"+战斗描写指南 |
- | 对话占比 | 无监控 | 新增维度31"对话占比监控",上限40% |
- | "话疗"检测 | 无 | 新增维度32"话疗模式检测",BOSS禁止被说服 |
- | 循环模式 | 无检测 | 新增维度33"叙事循环检测" |
- | 节奏评估 | 仅字数追踪 | 新增节奏质量评估(高潮密度、冲突强度) |
- | 最小审计集 | 5项 | 扩展为8项(新增战斗、对话、话疗、循环) |
- ---
- ## 核心定位
- 本 Skill 解决**八个**核心问题:
- 1. **长篇不跑偏**:通过 Truth Files 系统维护世界观、角色、伏笔、状态
- 2. **番茄能签约**:内置番茄平台规则、商业审查、付费转化检查
- 3. **AI没味道**:智能 AI 去味检查(替换而非删除),避免被平台限流
- 4. **会话连续性**:通过 handoff_current.md 实现跨会话无缝衔接
- 5. **规划能落地**:进度追踪面板,实时监控字数/章节偏离
- 6. **战斗有张力**:战斗场景专项审计,杜绝"话疗"解决BOSS
- 7. **对话不注水**:对话占比监控,防止全章变成聊天记录
- 8. **节奏不崩塌**:高潮密度追踪,防止后期注水和循环重复
- ---
- ## 核心工作流
- ```
- init -> write-next -> draft -> ai-detox -> audit -> revise -> extract-state -> state-update -> handoff-update
- ```
- ### v2.0.1 工作流强化
- ```
- 写前(强制5项):
- |- 读取 current_state.md -> 确认地点/时间/状态
- |- 读取 outline.md -> 确认本章目标、字数、是否高潮章
- |- 读取 pending_hooks.md -> 确认待处理伏笔
- |- 检查 battle-scene-guide.md -> 如是战斗章,加载战斗描写清单
- |- 检查 dialogue-monitor.md -> 确认当前对话占比趋势
- 写后(强制8项 -- 最小审计集 v2.0.1):
- |- 1. 字数检查(2000-2500字)
- |- 2. AI去味(智能替换,非机械删除)
- |- 3. 因果链检查(维度29,无跳脱)
- |- 4. 行为-规则一致性检查(维度28)
- |- 5. 时间线检查(无黑洞)
- |- 6. 战斗场景检查(维度30,高潮章必查) [新增]
- |- 7. 对话占比检查(维度31,上限40%) [新增]
- |- 8. 话疗模式检查(维度32,BOSS禁止被说服) [新增]
- 状态更新(强制):
- |- current_state.md
- |- pending_hooks.md
- |- handoff_current.md
- |- progress_tracker.md
- |- pattern-detection.md [新增,循环模式追踪]
- ```
- ---
- ## 项目结构
- 初始化后的番茄小说项目包含以下文件:
- ### 核心文件(必维护)
- | 文件 | 用途 | 更新频率 |
- |------|------|---------|
- | `story_bible.md` | 世界观、设定、势力、力量体系 | 新设定出现时 |
- | `book_rules.md` | 番茄平台规则 + 主角锁定 + 禁忌 | 项目开始时 |
- | `outline.md` | 分卷大纲、章节目标、payoff计划 | 每卷结束时回顾 |
- | `current_state.md` | 当前世界状态(最重要) | 每章更新 |
- | `pending_hooks.md` | 伏笔追踪、悬念管理 | 每章更新 |
- | `handoff_current.md` | 会话交接文件(活的文件) | 每章更新 |
- | `progress_tracker.md` | 进度追踪面板 | 每章更新 |
- | `pattern-detection.md` | 循环模式检测面板(v2.0.1新增) | 每5章更新 |
- ### 支撑文件(建议维护)
- | 文件 | 用途 |
- |------|------|
- | `chapter_summaries.md` | 每章摘要和状态变化 |
- | `character_matrix.md` | 角色关系、信任/冲突、信息边界 |
- | `character_knowledge.md` | POV管理、角色知识边界 |
- | `emotional_arcs.md` | 情感弧线追踪 |
- | `subplot_board.md` | 支线进度管理 |
- | `continuity_issues.md` | 已知问题追踪 |
- ### 风格文件
- | 文件 | 用途 |
- |------|------|
- | `style_guide.md` | 风格指南(含AI去味规则) |
- | `style_profile.json` | 量化风格统计 |
- ---
- ## 最小审计集(v2.0.1 核心改进)
- 每章写完后**必须执行**以下8项检查,不可跳过:
- ### 检查1-5:继承自 v1.1.1
- 详见 v1.1.1 文档。
- ### 检查6:战斗场景检查(维度30,v2.0.1新增)
- ```
- 触发条件:
- - 章节标题含"战斗/决战/对抗/最终/击杀"等关键词
- - outline.md 标记为"高潮章"
- - 本章出现BOSS级敌人
- 检查项:
- 1. 是否有实质性攻击/防御动作?(至少3次)
- 2. 是否通过"说服/理解/感化"解决冲突? -> 禁止!
- 3. 战斗意象是否重复?
- 4. 是否有战术博弈?(而非纯力量对轰)
- 5. 是否有代价/牺牲?
- 评级标准:
- PASS 优秀:有战术、有代价、意象丰富
- WARN 及格:有战斗但模式化
- FAIL 不合格:无战斗或"话疗"解决
- ```
- 详见:`references/battle-scene-guide.md`
- ### 检查7:对话占比检查(维度31,v2.0.1新增)
- ```
- 计算方法:
- 对话行数 = 统计引号内内容行数
- 总行数 = 章节总行数
- 对话占比 = 对话行数 / 总行数
- 标准:
- PASS 优秀:<= 25%(氛围/叙事为主)
- PASS 良好:26-35%(平衡得当)
- WARN 偏高:36-40%(需关注)
- FAIL 超标:> 40%(必须压缩对话)
- 趋势监控:
- - 记录每章对话占比
- - 若连续3章占比上升 -> 预警
- - 若占比 > 50% -> 立即报警
- ```
- 详见:`references/dialogue-monitor.md`
- ### 检查8:话疗模式检查(维度32,v2.0.1新增)
- ```
- 检测关键词:
- - "我理解你"
- - "说服"/"感化"/"唤醒"
- - "不是消灭,是救赎"
- - "融入"/"共存"/"和解"
- - BOSS被对话解决(三句话内放弃抵抗)
- 检查规则:
- 1. 普通敌人:可通过对话解决(限1次/卷)
- 2. 精英敌人:对话解决需铺垫3章+
- 3. BOSS级敌人:禁止通过对话解决!必须战斗!
- ```
- ### 检查9:循环模式检查(维度33,每5章执行)
- ```
- 检测模式:
- 1. "遇到新敌人 -> 发现苦衷 -> 对话解决 -> 获得力量"是否重复?
- 2. 连续3个章节结构是否高度相似?
- 3. 高潮章是否真的有高潮?
- 预警规则:
- - 同模式出现2次 -> WARN 提醒
- - 同模式出现3次+ -> FAIL 报警,要求变化
- ```
- 详见:`references/pattern-detection.md`
- ---
- ## 进度追踪面板(v2.0.1升级)
- ### 新增:节奏质量评估
- ```
- 高潮密度:
- 高潮章数量 / 总章数
- 标准:>= 15%(每7章至少1个高潮)
- 冲突强度趋势:
- 统计每章冲突解决方式
- 战斗解决占比应 >= 对话解决占比
- 节奏曲线:
- 绘制章节质量曲线
- 检测是否"前高后低"崩塌
- ```
- 详见:`assets/project-template/progress_tracker.md`
- ---
- ## 会话连续性工作流
- ### 新会话启动流程
- ```
- 1. 用户请求继续写作
- 2. AI请求加载 handoff_current.md
- 3. 用户提供文件内容(或AI从项目文件夹读取)
- 4. AI确认加载成功,报告当前进度
- 5. 继续写作或审计
- ```
- ### 会话交接协议
- **AI启动时必须说**:
- > "请提供 handoff_current.md 文件内容,我将加载当前进度。"
- **加载成功后必须确认**:
- > "已加载 handoff_current.md,当前进度:ch{{N}},下一章:ch{{N+1}}。
- > 主角当前在[地点],持有[物品],待处理伏笔[X个]。
- > 总进度:{{actual}}字/{{target}}字({{percent}}%)。
- > 对话占比趋势:{{trend}}。上次高潮章:ch{{last_climax}}。"
- ---
- ## 番茄专用功能
- ### 1. 平台红线检查
- 详见:`references/fanqie-platform-rules.md`
- ### 2. 生死线检查
- 前三章必须通过,详见:`references/commercial-review.md`
- ### 3. AI去味专项检查
- 详见:`references/ai-detox-guide.md`
- ### 4. 商业审查
- 详见:`references/commercial-review.md`
- ### 5. 战斗场景指南(v2.0.1新增)
- 详见:`references/battle-scene-guide.md`
- ### 6. 对话占比监控(v2.0.1新增)
- 详见:`references/dialogue-monitor.md`
- ---
- ## 教训库(v2.0.1)
- > 以下是从大量长篇网文实战创作与深度审阅中总结的典型问题模式,必须在审计时特别关注。
- ### 问题1-8:继承自 v1.1.1
- (略,见 v1.1.1 文档)
- ### 问题9:战斗描写崩溃(v2.0.1新增,严重)
- - **表现**:中后期战斗场景逐渐消失,被大段对话取代
- - **根因**:无战斗场景专项检查,主题表达压过情节张力
- - **预防**:维度30强制检查,高潮章必须有实质性战斗描写
- ### 问题10:对话占比失控(v2.0.1新增,严重)
- - **表现**:对话占比随篇幅推进持续飙升,后期章节近乎对话体
- - **根因**:无对话占比监控,后期写作倾向用对话凑字数
- - **预防**:维度31持续监控,上限40%,超标必须压缩
- ### 问题11:"话疗"模式泛滥(v2.0.1新增,严重)
- - **表现**:多个重要敌人被对话/说服解决,缺乏正当战斗
- - **根因**:无"话疗"检测机制,主题表达滥用为万能解法
- - **预防**:维度32检测,BOSS级敌人禁止被说服
- ### 问题12:循环模式重复(v2.0.1新增,中等)
- - **表现**:固定叙事循环(如"觉醒-升级-新敌人")无变化反复出现
- - **根因**:无循环模式检测,写作陷入套路化
- - **预防**:维度33检测,同模式出现3次以上强制报警
- ### 问题13:反派降智(v2.0.1新增,严重)
- - **表现**:终极BOSS逻辑崩塌,轻易被主角说服或击败
- - **根因**:急于表达主题,牺牲反派行为逻辑与角色尊严
- - **预防**:BOSS必须有独立且自洽的逻辑体系,主角需找到逻辑漏洞而非简单"理解"
- ### 问题14:节奏崩塌(v2.0.1新增,严重)
- - **表现**:作品前半部质量稳定,后半部出现灾难性下滑
- - **根因**:无节奏质量评估机制,后期注水赶工
- - **预防**:节奏曲线持续监控,"前高后低"趋势及时报警
- ---
- ## 版本历史
- ### v2.0.1(2026-05-30)
- - **重大升级**:基于长篇实战经验与深度审阅的全面重构
- - **新增维度30**:战斗场景审计(高潮章必查)
- - **新增维度31**:对话占比监控(上限40%)
- - **新增维度32**:话疗模式检测(BOSS禁止被说服)
- - **新增维度33**:叙事循环检测(同模式3次+报警)
- - **新增文件**:battle-scene-guide.md、dialogue-monitor.md、pattern-detection.md
- - **最小审计集扩展**:5项 -> 8项
- - **进度追踪升级**:新增节奏质量评估、高潮密度、冲突强度趋势
- - **教训库扩展**:新增问题9-14
- ### v1.1.1(2026-05-14)
- - 新增:progress_tracker.md、维度29、最小审计集、智能替换表
- - 修复:规划执行脱节、AI去味破坏语义、情节跳脱无因果
- ### v1.1.0
- - 新增:handoff_current.md、维度28、教训库
- ### v1.0.0
- - 初始版本
README.md · README · 292 lines
View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)
- <p align="center">
- <h1>🌱 Fanqie Novel Skill</h1>
- <p><strong>Production-Grade Writing System for Fanqie (Tomato) Novel Platform</strong></p>
- <p>
- <a href="#features">✨ Features</a> •
- <a href="#quick-start">🚀 Quick Start</a> •
- <a href="#architecture">🏗️ Architecture</a> •
- <a href="#changelog">📝 Changelog</a> •
- <a href="#license">📄 License</a>
- </p>
- </p>
- ---
- ## [中文介绍](#中文介绍) | [English](#english)
- ---
- <a id="中文介绍"></a>
- ## 🇨🇳 中文介绍
- ### 这是什么?
- Fanqie Novel Skill 是一个专为**番茄小说平台**设计的 AI 辅助长篇网文写作系统。它提供了一套完整的工作流,帮助你在番茄平台上创作出高质量、可签约、能变现的长篇小说。
- ### 为什么需要它?
- AI 辅助写作在番茄平台面临三大挑战:
- | 挑战 | 后果 | 本 Skill 的解决方案 |
- |------|------|-------------------|
- | AI 味被平台限流 | 作品被降权或下架 | 智能去味系统(替换而非删除) |
- | 长篇规划执行脱节 | 规划 120 万字实际只写 70 万 | 进度追踪面板 + 预警系统 |
- | 审计维度太多导致空转 | 28 个维度一个都不执行 | 最小审计集(8 项强制) |
- | 后期质量崩塌 | 前期精彩后期注水 | 节奏质量评估 + 循环模式检测 |
- | "话疗"解决冲突 | BOSS 被三句话说服 | 战斗场景审计 + 话疗模式检测 |
- ### 核心特性
- - **🛡️ AI 去味系统** — 智能替换表,语义完整性优先,避免机械删除
- - **📊 进度追踪面板** — 实时监控字数/章节偏离,预警规则自动触发
- - **🔍 最小审计集(8 项强制)** — 每章必查,不执行不通过
- - **⚔️ 战斗场景审计** — 杜绝"话疗"解决 BOSS,确保战斗有张力
- - **💬 对话占比监控** — 上限 40%,防止全章变成聊天记录
- - **🔄 循环模式检测** — 检测叙事模式重复,保持新鲜感
- - **📈 节奏质量评估** — 高潮密度、冲突强度趋势、节奏曲线
- - **🔗 会话连续性** — 跨会话无缝衔接,支持超长篇创作
- ### 工作流
- ```
- 初始化 → 读取状态 → 写作 → AI去味 → 审计 → 修订 → 状态更新 → 会话交接
- ```
- 每章写作前后都有强制检查点,确保质量不滑坡。
- ---
- <a id="english"></a>
- ## 🇺🇸 English
- ### What is this?
- Fanqie Novel Skill is a production-grade AI-assisted writing system designed specifically for the **Fanqie (Tomato) Novel Platform**. It provides a complete workflow for creating high-quality, publishable long-form web novels.
- ### Why do you need it?
- AI-assisted writing on the Fanqie platform faces critical challenges:
- | Challenge | Consequence | Solution |
- |-----------|-------------|----------|
- | AI-flavored text gets throttled | Works get deprioritized or removed | Smart de-flavoring system (replace, don't delete) |
- | Long-form planning drift | Plan 1.2M words, deliver 700K | Progress tracking dashboard + alerts |
- | Too many audit dimensions = none executed | 28 dimensions, zero compliance | Minimum audit set (8 mandatory checks) |
- | Quality collapse in later chapters | Strong start, weak finish | Rhythm quality assessment + pattern detection |
- | Conflicts resolved by "talking therapy" | Boss defeated by 3 sentences | Battle scene audit + talk-therapy detection |
- ### Core Features
- - **🛡️ AI De-flavoring** — Smart replacement table, semantic integrity first
- - **📊 Progress Tracking** — Real-time word/chapter monitoring with auto-alerts
- - **🔍 Minimum Audit Set** — 8 mandatory checks per chapter
- - **⚔️ Battle Scene Audit** — No "talk-therapy" boss defeats, ensure combat tension
- - **💬 Dialogue Ratio Monitor** — 40% cap, prevent chapters becoming chat logs
- - **🔄 Pattern Detection** — Detect narrative loop repetition, maintain freshness
- - **📈 Rhythm Quality Assessment** — Climax density, conflict intensity trends
- - **🔗 Session Continuity** — Seamless cross-session handoff for ultra-long novels
- ### Workflow
- ```
- Init → Read State → Write → De-flavor → Audit → Revise → State Update → Handoff
- ```
- Mandatory checkpoints before and after each chapter ensure consistent quality.
- ---
- ## Features
- ### 🛡️ AI De-flavoring System
- Not just a banned-word list. A smart replacement engine that:
- - **Replaces, not deletes** — Preserves metaphors and figurative language
- - **Context-aware** — Allows banned words in dialogue (characters can sound AI-like, narration can't)
- - **Semantic verification** — Post-replacement read-aloud test
- - **9-word smart replacement table** with context-specific alternatives
- ### 📊 Progress Tracking Dashboard
- - Total/volume/chapter progress tracking
- - **Rhythm quality metrics** (v2.0): climax density, conflict intensity trends, quality curve
- - **12 auto-alert rules**: word count deviation, dialogue overflow, pattern repetition, quality collapse
- - Quality scoring system (per-chapter + full-book)
- ### 🔍 Minimum Audit Set (8 Mandatory Checks)
- | # | Check | Frequency |
- |---|-------|-----------|
- | 1 | Word count (2000-2500) | Every chapter |
- | 2 | AI de-flavoring | Every chapter |
- | 3 | Causal chain | Every chapter |
- | 4 | Behavior-rule consistency | Every chapter |
- | 5 | Timeline continuity | Every chapter |
- | 6 | Battle scene quality | Combat chapters |
- | 7 | Dialogue ratio (≤40%) | Every chapter |
- | 8 | Talk-therapy detection | Every chapter |
- Plus: **Pattern detection** every 5 chapters, **Full audit** (33 dimensions) every 10 chapters.
- ### ⚔️ Battle Scene System (v2.0)
- - **Boss battles must be fought, not talked through**
- - Combat checklist: ≥3 attack actions, tactical gameplay, cost/sacrifice
- - Battle imagery diversity check (prevent "golden light" repetition)
- - Full battle scene writing guide with templates
- ### 🔄 Pattern Detection (v2.0)
- - Detects narrative loop repetition (same pattern 3+ times = alert)
- - Chapter structure similarity analysis
- - Climax chapter authenticity verification
- ---
- ## Quick Start
- ### Prerequisites
- - AI coding assistant with skill support (e.g., Claude Code, Cursor)
- - Fanqie (Tomato) Novel Platform account
- ### Installation
- 1. Download or clone this repository
- 2. Place the skill folder in your assistant's skill directory
- 3. Initialize a new novel project using the provided templates
- ### Project Structure
- After initialization, your novel project contains:
- ```
- your-novel/
- ├── story_bible.md # World-building, settings, power system
- ├── book_rules.md # Platform rules + protagonist lock + taboos
- ├── outline.md # Volume outlines, chapter goals, payoff plans
- ├── current_state.md # Current world state (most important!)
- ├── pending_hooks.md # Foreshadowing & suspense tracking
- ├── handoff_current.md # Session handoff file (living document)
- ├── progress_tracker.md # Progress + rhythm quality dashboard
- ├── pattern_detection.md # Narrative pattern tracking
- ├── chapter_summaries.md # Per-chapter summaries
- ├── character_matrix.md # Character relationships & knowledge boundaries
- ├── style_guide.md # Style guide (with AI de-flavoring rules)
- └── chapters/
- └── ch001.md
- ```
- ### Per-Chapter Workflow
- ```
- BEFORE writing (5 mandatory):
- ├─ Read current_state.md → Confirm location/time/status
- ├─ Read outline.md → Confirm chapter goal, word count, is-climax
- ├─ Read pending_hooks.md → Check unresolved foreshadowing
- ├─ Check battle-scene-guide.md → If combat chapter, load checklist
- └─ Check dialogue-monitor.md → Confirm dialogue ratio trend
- AFTER writing (8 mandatory):
- ├─ 1. Word count check (2000-2500)
- ├─ 2. AI de-flavoring (smart replace)
- ├─ 3. Causal chain check
- ├─ 4. Behavior-rule consistency
- ├─ 5. Timeline continuity
- ├─ 6. Battle scene check (combat chapters)
- ├─ 7. Dialogue ratio check (≤40%)
- └─ 8. Talk-therapy detection (bosses can't be talked down)
- STATE UPDATE (mandatory):
- ├─ current_state.md
- ├─ pending_hooks.md
- ├─ handoff_current.md
- ├─ progress_tracker.md
- └─ pattern_detection.md
- ```
- ---
- ## Architecture
- ### File Organization
- ```
- fanqie-novel-skill/
- ├── SKILL.md # Main skill definition
- ├── README.md # This file
- ├── CHANGELOG.md # Version history
- ├── VERSION # Current version
- ├── assets/
- │ └── project-template/ # Novel project templates
- │ ├── story_bible.md
- │ ├── book_rules.md
- │ ├── outline.md
- │ ├── current_state.md
- │ ├── pending_hooks.md
- │ ├── handoff_current.md
- │ ├── progress_tracker.md
- │ └── ...
- ├── examples/
- │ └── demo-novel/ # Example novel project
- └── references/ # Reference guides
- ├── ai-detox-guide.md # AI de-flavoring guide
- ├── audit-dimensions.md # 33 audit dimensions
- ├── battle-scene-guide.md # Battle writing guide
- ├── dialogue-monitor.md # Dialogue ratio monitoring
- ├── pattern-detection.md # Narrative pattern detection
- ├── commercial-review.md # Commercial viability checks
- └── fanqie-platform-rules.md # Platform rules & red lines
- ```
- ### Audit Dimensions (33 Total)
- | Category | Dimensions | Key Focus |
- |----------|:---:|-----------|
- | Minimum Set (Mandatory) | 8 | Word count, AI flavor, causality, timeline, battle, dialogue, talk-therapy |
- | Core | 10 | Character consistency, timeline, world-building, logic, pacing |
- | Style & Anti-formula | 6 | Repetition, show-don't-tell, AI transitions |
- | Structure | 4 | Outline drift, subplot stagnation, payoff |
- | Fanqie-specific | 5 | Platform red lines, golden 3 chapters, commercial checks |
- ---
- ## Changelog
- See [CHANGELOG.md](CHANGELOG.md) for detailed version history.
- ### Highlights
- - **v2.0.1** — Major upgrade: battle scene audit, dialogue monitoring, talk-therapy detection, pattern detection, rhythm quality assessment
- - **v1.1.1** — Progress tracking, causal chain check, minimum audit set, smart AI de-flavoring
- - **v1.1.0** — Session handoff, behavior-rule consistency
- - **v1.0.0** — Initial release
- ---
- ## Contributing
- Contributions are welcome! Please feel free to submit a Pull Request.
- 1. Fork the repository
- 2. Create your feature branch (`git checkout -b feature/amazing-feature`)
- 3. Commit your changes (`git commit -m 'Add amazing feature'`)
- 4. Push to the branch (`git push origin feature/amazing-feature`)
- 5. Open a Pull Request
- ---
- ## License
- This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.
- ---
- ## Author
- **Jayce** — Web novel enthusiast & AI writing system designer
- <p align="center">
- If you find this skill helpful, please consider giving it a ⭐ on GitHub!
- </p>
Ratings and comments
Sign in to rate this skill.
Sign inFree to use. If it’s helped, pass it on.