Best use case
小说写作技能包 (novel-plus) is best used when you need a repeatable AI agent workflow instead of a one-off prompt. It is especially useful for teams working in multi. ## 版本
## 版本
Users should expect a more consistent workflow output, faster repeated execution, and less time spent rewriting prompts from scratch.
Practical example
Example input
Use the "小说写作技能包 (novel-plus)" skill to help with this workflow task. Context: ## 版本
Example output
A structured workflow result with clearer steps, more consistent formatting, and an output that is easier to reuse in the next run.
When to use this skill
- Use this skill when you want a reusable workflow rather than writing the same prompt again and again.
When not to use this skill
- Do not use this when you only need a one-off answer and do not need a reusable workflow.
- Do not use it if you cannot install or maintain the related files, repository context, or supporting tools.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/novel-plus/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How 小说写作技能包 (novel-plus) Compares
| Feature / Agent | 小说写作技能包 (novel-plus) | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
## 版本
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
Related Guides
AI Agents for Marketing
Discover AI agents for marketing workflows, from SEO and content production to campaign research, outreach, and analytics.
AI Agents for Startups
Explore AI agent skills for startup validation, product research, growth experiments, documentation, and fast execution with small teams.
AI Agents for Coding
Browse AI agent skills for coding, debugging, testing, refactoring, code review, and developer workflows across Claude, Cursor, and Codex.
SKILL.md Source
# 小说写作技能包 (novel-plus)
## 版本
**v2.4** - 多智能体协作版
## 描述
这是一个完整的小说写作技能包,采用 **多智能体(Multi-Agent)协作架构**。主控(Coordinator)统一调度各专业 Agent,覆盖项目初始化、世界观构建、角色圣经、大纲细纲、章节协作写作、审核迭代全生命周期。
## 核心特性
- **多 Agent 协作**:12 个专业 Agent,各司其职
- **自动注册**:Phase 0 初始化时自动注册所有 Agent
- **智能调度**:主控根据任务类型自动分发
- **自动推进**:支持连续写作 n 章,减少人工干预
- **完整存档**:版本化管理,可追溯
## 激活关键词
| 关键词 | 激活阶段 |
|--------|----------|
| "新建小说" / "新项目" | Phase 0 |
| "开始写小说" / "创建小说" | Phase 0(引导) |
| "世界观" / "设定" | Phase 1 Step 1 |
| "角色" / "人物" | Phase 1 Step 2 |
| "大纲" / "章节规划" | Phase 1 Step 3-4 |
| "审核大纲" / "检查架构" | Phase 1 Step 5 |
| "写第X章" / "写第X到Y章" | Phase 2 |
| "写第X章,自动推进N章" | Phase 2(自动推进) |
| "读者反馈" | Phase 2(立即触发读者模拟) |
| "继续写作" / "停止写作" | Phase 2(恢复/停止自动推进) |
| "重写第X章" / "重写第X章第Y段" | Phase 3 |
| "修改" / "补设定" | Phase 3 |
## 目录结构
```
novel-plus/
├── manifest.json # 技能包元数据配置(含Agent注册信息)
├── instructions.md # 完整技能指令(v2.4)
├── AGENT_SYSTEM.md # Agent系统指南
├── README.md # 本说明文件
├── agents/ # Agent定义目录
│ ├── coordinator.md # 主控协调
│ ├── worldbuilder.md # 世界观构建
│ ├── character-designer.md # 角色设计
│ ├── outline-planner.md # 大纲规划
│ ├── chapter-outliner.md # 章节细纲
│ ├── main-writer.md # 主笔
│ ├── ooc-guardian.md # OOC守护
│ ├── battle-agent.md # 战斗写作
│ ├── final-reviewer.md # 终稿审核
│ ├── reader-simulator.md # 读者模拟
│ ├── style-anchor-generator.md # 风格锚定生成
│ └── rolling-summarizer.md # 滚动摘要
├── references/ # 参考模板目录
│ └── ...
└── templates/ # 项目模板目录
└── project-template/ # 完整项目模板
```
## Agent 架构
```
┌─────────────────┐
│ 用户 │
└────────┬────────┘
│
▼
┌─────────────────────────┐
│ Coordinator (主控) │
│ MiniMax M2.7 │
│ - 任务调度 │
│ - 存档管理 │
│ - 用户交互 │
└───────────┬─────────────┘
│
┌──────────┬──────────┼──────────┬──────────┐
▼ ▼ ▼ ▼ ▼
┌─────────┐┌─────────┐┌─────────┐┌─────────┐┌─────────┐
│World- ││Character││Outline ││Chapter ││ Main │
│builder ││Designer ││Planner ││Outliner ││ Writer │
│(Opus) ││(Opus) ││(Opus) ││(Kimi) ││(Kimi) │
└─────────┘└─────────┘└─────────┘└─────────┘└─────────┘
│ │ │ │ │
└──────────┴──────────┴──────────┴──────────┘
│
┌──────────┴──────────┐
▼ ▼
┌─────────┐ ┌─────────┐
│ Battle │ │ Final │
│ Agent │ │Reviewer │
│(Gemini) │ │ (Opus) │
└─────────┘ └─────────┘
```
## 使用方法
### 作为 OpenClaw Skill 使用
1. 将此目录复制到 OpenClaw 的 skills 目录
2. 在 OpenClaw 配置中注册此 skill
3. 使用激活关键词(如"新建小说")开始对话
## 项目目录结构
```
projects/<项目名>/
├── meta/
│ ├── project.md # 项目元数据
│ ├── metadata.json # 统一元数据
│ ├── config.md # 模型分工配置
│ ├── style-anchor.md # 风格锚定文档
│ └── agent-registry.json # Agent注册表(v2.4新增)
├── worldbuilding/
│ └── world.md # 世界观文档
├── characters/
│ ├── protagonist.md # 主角档案
│ └── characters.md # 其他角色
├── outline/
│ ├── outline.md # 全书大纲
│ └── chapter-outline.md # 章节细纲
├── chapters/
│ ├── ch01.md # 当前最新版
│ └── history/ # 历史版本
├── archive/
│ ├── archive.md # 项目日志
│ └── reader-feedback/ # 读者反馈
└── references/
└── ...
```
## 生命周期阶段
```
Phase 0 ──► Phase 1 ──► Phase 2 ──► Phase 3
│ │ │ │
初始化 前期架构 章节写作 维护迭代
```
### Phase 0:新建项目初始化
- 收集核心设定
- 创建项目目录树
- **自动注册所有 Agent 到 agent-registry.json**
- 初始化元数据文件
### Phase 1:前期架构阶段
- **Worldbuilder** 构建世界观
- **CharacterDesigner** 创建角色圣经
- **OutlinePlanner** 设计全书大纲
- **ChapterOutliner** 生成章节细纲
- **StyleAnchorGenerator** 生成风格锚定文档
### Phase 2:章节写作阶段
- **MainWriter** 主笔写作
- **OOCGuardian** 条件触发一致性检查
- **BattleAgent** 高强度战斗场面
- **FinalReviewer** 重点章节终审
- **RollingSummarizer** 每5-10章更新摘要
- **ReaderSimulator** 每10章读者反馈
### Phase 3:维护与迭代
- 章节/段落重写
- 补设定/世界观扩展
- 修改大纲
- 风格调整
## Agent 职责表
| Agent | 模型 | 职责 | 调用时机 |
|-------|------|------|----------|
| Coordinator | M2.7 | 总控协调 | 始终运行 |
| Worldbuilder | Opus | 世界观构建 | Phase 1 |
| CharacterDesigner | Opus | 角色设计 | Phase 1 |
| OutlinePlanner | Opus | 大纲规划 | Phase 1 |
| ChapterOutliner | Kimi | 章节细纲 | Phase 1 |
| MainWriter | Kimi | 初稿+润色 | Phase 2 |
| OOCGuardian | Qwen | 一致性检查 | 条件触发 |
| BattleAgent | Gemini | 高强度战斗 | 条件触发 |
| FinalReviewer | Opus | 终审 | 重点章节 |
| ReaderSimulator | Doubao | 读者反馈 | 每10章 |
| StyleAnchorGenerator | Opus | 风格文档 | Phase 1 |
| RollingSummarizer | Qwen | 滚动摘要 | 每5-10章 |
## 核心铁律(优先级降序)
1. **P0 用户边界违规**(绝对禁止)
2. **P1 人设崩坏/OOC**(严格禁止)
3. **P2 设定逻辑冲突**(必须修复)
4. **P3 风格偏离**(应当修复)
5. **P4 节奏字数偏差**(优化项)
## 自动推进参数
| 参数 | 默认值 | 说明 |
|------|--------|------|
| auto_advance_chapters | 4 | 一次自动写4章 |
| write_interval_seconds | 6 | 每章完成后等待秒数 |
| auto_confirm | false | 是否自动确认继续 |
## 文档说明
| 文档 | 说明 |
|------|------|
| `manifest.json` | 技能包配置,包含 Agent 注册信息 |
| `instructions.md` | 完整技能指令(详细流程) |
| `AGENT_SYSTEM.md` | Agent 系统指南(调度机制、回退策略) |
| `agents/*.md` | 各 Agent 的定义和 Prompt |
| `references/*.md` | 模板文件 |
| `templates/` | 项目模板 |
## 版本历史
| 版本 | 日期 | 变更 |
|------|------|------|
| v2.0 | 2026-03-21 | 初始Phase架构版 |
| v2.1 | 2026-03-21 | 优化读者模拟、滚动摘要等 |
| v2.2 | 2026-03-21 | 新增自动推进机制 |
| v2.3 | 2026-03-21 | 优化常规章节调用次数 |
| **v2.4** | 2026-03-22 | **新增多Agent协作系统,自动注册机制** |Related Skills
12agent-novel
中文长篇小说多智能体创作体系(12Agent)。适用于新建长篇小说项目、搭建世界观与大纲、逐章写作、自动推进与读者反馈等长流程创作任务;不适用于短篇、诗歌、散文、翻译或非小说写作。
novelai-openclaw-adaptor
Explain how to connect NovelAI to OpenClaw through a local OpenAI-compatible shim. Use when the user wants configuration guidance for a local NovelAI adaptor, model selection, or OpenClaw `base_url` setup.
weather-plus
获取全球城市天气信息和预报。使用场景包括查询当前天气、未来7天预报、空气质量、穿衣建议等。适用于用户问天气、温度、预报等。
mockplus-reader
读取和分析 MockPlus 在线设计页面。用于:(1)打开并解析 MockPlus 网页链接,(2)提取页面中的设计信息、结构、组件,(3)分析原型稿内容和交互说明。当用户发送 MockPlus 链接或要求分析原型稿时使用此技能。
novel-free
中文长篇小说多智能体创作技能(v1.0.1)。支持从零开始新建长篇小说项目,完成世界观/角色/大纲/章节写作、自动推进与读者反馈的完整流程。内置固定层压缩、fixed-context.md强制缓存、摘要+OOC合并触发三项核心优化,新增用户体验优化和自动化工具,显著降低token消耗与操作复杂度。适用场景:新建长篇小说、续写/批量推进章节、自动推进、重写/维护迭代。不适用于短篇(<1万字)、诗歌、散文、翻译或非小说写作。
novel-forge
Long-form novel workflow for creating, continuing, resuming, and repairing serialized fiction with externalized project state, role-to-model mapping, worldbuilding, character sheets, full outlines, 10-chapter batch outlines, style sampling, chapter drafting, consistency review, memory tracking, and spawned multi-session collaboration. Use when the user asks to start a novel project, continue or resume a draft, recover from truncation, assign models to roles, generate canon or chapters, review for consistency, or maintain a long-running fiction project across many chapters. Supports single-agent or multi-agent execution, with multi-agent as the default; when multi-agent is selected, first surface the available model inventory and the novel-writing role list, then ask the user for an explicit role→model mapping before any canon work. Once the user has provided the mapping, persist it in project state and drive stage work with `sessions_spawn` using the mapped roles rather than treating the mapping as passive metadata. The main session may only create the project shell and route work; it must not author canon files.
---
name: article-factory-wechat
humanizer
Remove signs of AI-generated writing from text. Use when editing or reviewing text to make it sound more natural and human-written. Based on Wikipedia's comprehensive "Signs of AI writing" guide. Detects and fixes patterns including: inflated symbolism, promotional language, superficial -ing analyses, vague attributions, em dash overuse, rule of three, AI vocabulary words, negative parallelisms, and excessive conjunctive phrases.
find-skills
Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skill should be used when the user is looking for functionality that might exist as an installable skill.
tavily-search
Use Tavily API for real-time web search and content extraction. Use when: user needs real-time web search results, research, or current information from the web. Requires Tavily API key.
baidu-search
Search the web using Baidu AI Search Engine (BDSE). Use for live information, documentation, or research topics.
agent-autonomy-kit
Stop waiting for prompts. Keep working.