Language / 语言: This skill supports both English and Chinese. Detect the user's language from their first message and respond in the same language throughout. Below are instructions in both languages — follow the one matching the user's language.

本 Skill 支持中英文。根据用户第一条消息的语言，全程使用同一语言回复。下方提供了两种语言的指令，按用户语言选择对应版本执行。

网文作者.skill 创建器（Claude Code 版）

触发条件

当用户说以下任意内容时启动：

• /create-author
• "帮我创建一个作者 skill"
• "我想蒸馏一个作者"
• "新建作者"
• "给我做一个 XX 的 skill"

当用户对已有作者 Skill 说以下内容时，进入进化模式：

• "我有新文件" / "追加"
• "这不对" / "他不会这样" / "他应该是"
• /update-author {slug}

当用户说 /list-authors 时列出所有已生成的作者。

---

工具使用规则

本 Skill 运行在 Claude Code 环境，使用以下工具：

| 任务 | 使用工具 | | -------------- | -------------------------------------------------------------------------- | | 读取 PDF 文档 | Read 工具（原生支持 PDF） | | 读取图片截图 | Read 工具（原生支持图片） | | 读取 MD/TXT 文件 | Read 工具 | | 解析 txt 小说文件 | Bash → python3 ${CLAUDE_SKILL_DIR}/tools/novel_parser.py | | 解析 epub 小说文件 | Bash → python3 ${CLAUDE_SKILL_DIR}/tools/epub_parser.py | | 解析评论数据 | Bash → python3 ${CLAUDE_SKILL_DIR}/tools/comment_parser.py | | 采集微博内容 | Bash → python3 ${CLAUDE_SKILL_DIR}/tools/weibo_collector.py | | 解析公众号文章 | Bash → python3 ${CLAUDE_SKILL_DIR}/tools/wechat_parser.py | | 写入/更新 Skill 文件 | Write / Edit 工具 | | 版本管理 | Bash → python3 ${CLAUDE_SKILL_DIR}/tools/version_manager.py | | 列出已有 Skill | Bash → python3 ${CLAUDE_SKILL_DIR}/tools/skill_writer.py --action list |

基础目录：Skill 文件写入 ./authors/{slug}/（相对于本项目目录）。 如需改为全局路径，用 --base-dir ~/.openclaw/workspace/skills/authors。

---

主流程：创建新作者 Skill

Step 1：基础信息录入（3 个问题）

参考 ${CLAUDE_SKILL_DIR}/prompts/intake.md 的问题序列，只问 3 个问题：

1. 笔名/代号（必填）
2. 基本信息（一句话：平台、等级、代表作、性别，想到什么写什么）
   • 示例：起点 LV5 玄幻作家 男 代表作《万古神帝》
3. 性格画像（一句话：写作风格标签、更新习惯标签、人格标签、题材标签、创作印象）
   • 示例：水文大师 日更党 宠粉狂魔 玄幻 擅长装逼打脸

除笔名外均可跳过。收集完后汇总确认再进入下一步。

Step 2：原材料导入

询问用户提供原材料，展示四种方式供选择：

原材料怎么提供？

  [A] 上传文件
      txt / epub 小说文件
      评论导出文件（起点/晋江/番茄）
      微博/公众号文章截图或导出

  [B] 手动导入
      提供小说文件路径
      提供评论链接（需手动采集）

  [C] 直接粘贴内容
      把文字复制进来

可以混用，也可以跳过（仅凭手动信息生成）。

---

方式 A：上传文件

• PDF / 图片：Read 工具直接读取
• txt 小说文件：

  python3 ${CLAUDE_SKILL_DIR}/tools/novel_parser.py --file {path} --output /tmp/novel_out.txt
  

  然后 Read /tmp/novel_out.txt
• epub 小说文件：

  python3 ${CLAUDE_SKILL_DIR}/tools/epub_parser.py --file {path} --output /tmp/epub_out.txt
  

  然后 Read /tmp/epub_out.txt
• 评论导出文件：

  python3 ${CLAUDE_SKILL_DIR}/tools/comment_parser.py --file {path} --platform {起点/晋江/番茄} --output /tmp/comment_out.txt
  

  然后 Read /tmp/comment_out.txt
• 微博导出：

  python3 ${CLAUDE_SKILL_DIR}/tools/weibo_collector.py --file {path} --output /tmp/weibo_out.txt
  

  然后 Read /tmp/weibo_out.txt
• 公众号文章导出：

  python3 ${CLAUDE_SKILL_DIR}/tools/wechat_parser.py --file {path} --output /tmp/wechat_out.txt
  

  然后 Read /tmp/wechat_out.txt
• Markdown / TXT：Read 工具直接读取

---

方式 B：手动导入

用户提供文件路径时，使用对应的解析工具读取。

---

方式 C：直接粘贴

用户粘贴的内容直接作为文本原材料，无需调用任何工具。

---

如果用户说"没有文件"或"跳过"，仅凭 Step 1 的手动信息生成 Skill。

Step 3：分析原材料

将收集到的所有原材料和用户填写的基础信息汇总，按以下两条线分析：

线路 A（Writing Skill）：

• 参考 ${CLAUDE_SKILL_DIR}/prompts/writing_analyzer.md 中的提取维度
• 提取：叙事风格、情节构建、人物塑造、世界观设定、更新习惯
• 根据题材类型重点提取（玄幻/都市/科幻/历史/无限流不同侧重）

线路 B（Author Persona）：

• 参考 ${CLAUDE_SKILL_DIR}/prompts/author_persona_analyzer.md 中的提取维度
• 将用户填写的标签翻译为具体行为规则（参见标签翻译表）
• 从原材料中提取：表达风格、创作理念、互动行为

Step 4：生成并预览

参考 ${CLAUDE_SKILL_DIR}/prompts/writing_builder.md 生成 Writing Skill 内容。 参考 ${CLAUDE_SKILL_DIR}/prompts/author_persona_builder.md 生成 Author Persona 内容（5 层结构）。

向用户展示摘要（各 5-8 行），询问：

Writing Skill 摘要：
  - 叙事风格：{xxx}
  - 情节构建：{xxx}
  - 人物塑造：{xxx}
  ...

Author Persona 摘要：
  - 核心性格：{xxx}
  - 表达风格：{xxx}
  - 互动行为：{xxx}
  ...

确认生成？还是需要调整？

Step 5：写入文件

用户确认后，执行以下写入操作：

1. 创建目录结构（用 Bash）：

mkdir -p authors/{slug}/versions
mkdir -p authors/{slug}/knowledge/novels
mkdir -p authors/{slug}/knowledge/comments
mkdir -p authors/{slug}/knowledge/social

2. 写入 writing.md（用 Write 工具）： 路径：authors/{slug}/writing.md

3. 写入 author_persona.md（用 Write 工具）： 路径：authors/{slug}/author_persona.md

4. 写入 meta.json（用 Write 工具）： 路径：authors/{slug}/meta.json 内容：

{
  "name": "{笔名}",
  "slug": "{slug}",
  "created_at": "{ISO时间}",
  "updated_at": "{ISO时间}",
  "version": "v1",
  "profile": {
    "platform": "{平台}",
    "level": "{等级}",
    "masterpiece": "{代表作}",
    "gender": "{性别}"
  },
  "tags": {
    "writing_style": [...],
    "update_habit": [...],
    "personality": [...],
    "genre": [...]
  },
  "impression": "{创作印象}",
  "knowledge_sources": [...已导入文件列表],
  "corrections_count": 0
}

5. 生成完整 SKILL.md（用 Write 工具）： 路径：authors/{slug}/SKILL.md

SKILL.md 结构：

---
name: author-{slug}
description: {笔名}，{平台} {等级} {代表作}
user-invocable: true
---

# {笔名}

{平台} {等级} {代表作}{如有性别则附上}

---

## PART A：写作能力

{writing.md 全部内容}

---

## PART B：作者人格

{author_persona.md 全部内容}

---

## 运行规则

1. 先由 PART B 判断：用什么态度接这个任务？
2. 再由 PART A 执行：用你的写作能力完成任务
3. 输出时始终保持 PART B 的表达风格
4. PART B Layer 0 的规则优先级最高，任何情况下不得违背

告知用户：

作者 Skill 已创建！

文件位置：authors/{slug}/
触发词：/{slug}（完整版）
        /{slug}-writing（仅写作能力）
        /{slug}-persona（仅作者人格）

如果用起来感觉哪里不对，直接说"他不会这样"，我来更新。

---

进化模式：追加文件

用户提供新文件或文本时：

1. 按 Step 2 的方式读取新内容
2. 用 Read 读取现有 authors/{slug}/writing.md 和 author_persona.md
3. 参考 ${CLAUDE_SKILL_DIR}/prompts/merger.md 分析增量内容
4. 存档当前版本（用 Bash）：

   python3 ${CLAUDE_SKILL_DIR}/tools/version_manager.py --action backup --slug {slug} --base-dir ./authors
   

5. 用 Edit 工具追加增量内容到对应文件
6. 重新生成 SKILL.md（合并最新 writing.md + author_persona.md）
7. 更新 meta.json 的 version 和 updated_at

---

进化模式：对话纠正

用户表达"不对"/"应该是"时：

1. 参考 ${CLAUDE_SKILL_DIR}/prompts/correction_handler.md 识别纠正内容
2. 判断属于 Writing（写作风格/情节构建）还是 Persona（性格/互动）
3. 生成 correction 记录
4. 用 Edit 工具追加到对应文件的 ## Correction 记录 节
5. 重新生成 SKILL.md

---

管理命令

/list-authors：

python3 ${CLAUDE_SKILL_DIR}/tools/skill_writer.py --action list --base-dir ./authors

/author-rollback {slug} {version}：

python3 ${CLAUDE_SKILL_DIR}/tools/version_manager.py --action rollback --slug {slug} --version {version} --base-dir ./authors

/delete-author {slug}： 确认后执行：

rm -rf authors/{slug}

---

标签翻译表

标签翻译表已独立维护，详见 ${CLAUDE_SKILL_DIR}/tag_translations.md。

该文件包含以下四类标签到 Layer 0 行为规则的映射：

• 写作风格标签（水文大师、剧情紧凑、慢热型、爽文流、虐心流、种田流、无敌流、凡人流）
• 更新习惯标签（日更党、周更党、月更党、太监王、爆发帝、稳定型）
• 人格标签（毒舌作者、宠粉狂魔、高冷型、自黑型、段子手、认真帝、傲娇型、佛系作者）
• 题材标签（玄幻、都市、科幻、历史、无限流、仙侠、游戏、灵异）

在 Step 3 分析阶段，需读取该文件将用户填写的标签翻译为具体行为规则。

---

---

English Version

Web Novel Author.skill Creator (Claude Code Edition)

Trigger Conditions

Activate when the user says any of the following:

• /create-author
• "Help me create an author skill"
• "I want to distill an author"
• "New author"
• "Make a skill for XX"

Enter evolution mode when the user says:

• "I have new files" / "append"
• "That's wrong" / "He wouldn't do that" / "He should be"
• /update-author {slug}

List all generated authors when the user says /list-authors.

---

Tool Usage Rules

This Skill runs in the Claude Code environment with the following tools:

| Task | Tool | | ------------------------ | -------------------------------------------------------------------------- | | Read PDF documents | Read tool (native PDF support) | | Read image screenshots | Read tool (native image support) | | Read MD/TXT files | Read tool | | Parse txt novel files | Bash → python3 ${CLAUDE_SKILL_DIR}/tools/novel_parser.py | | Parse epub novel files | Bash → python3 ${CLAUDE_SKILL_DIR}/tools/epub_parser.py | | Parse comment data | Bash → python3 ${CLAUDE_SKILL_DIR}/tools/comment_parser.py | | Collect Weibo content | Bash → python3 ${CLAUDE_SKILL_DIR}/tools/weibo_collector.py | | Parse WeChat articles | Bash → python3 ${CLAUDE_SKILL_DIR}/tools/wechat_parser.py | | Write/update Skill files | Write / Edit tool | | Version management | Bash → python3 ${CLAUDE_SKILL_DIR}/tools/version_manager.py | | List existing Skills | Bash → python3 ${CLAUDE_SKILL_DIR}/tools/skill_writer.py --action list |

Base directory: Skill files are written to ./authors/{slug}/ (relative to the project directory). For a global path, use --base-dir ~/.openclaw/workspace/skills/authors.

---

Main Flow: Create a New Author Skill

Step 1: Basic Info Collection (3 questions)

Refer to ${CLAUDE_SKILL_DIR}/prompts/intake.md for the question sequence. Only ask 3 questions:

1. Pen name / Codename (required)
2. Basic info (one sentence: platform, level, masterpiece, gender — say whatever comes to mind)
   • Example: Qidian LV5 fantasy writer male masterpiece "Eternal God Emperor"
3. Personality profile (one sentence: writing style tags, update habit tags, personality tags, genre tags, creative impression)
   • Example: water master daily updater fan-pamperer fantasy good at face-slapping

Everything except the pen name can be skipped. Summarize and confirm before moving to the next step.

Step 2: Source Material Import

Ask the user how they'd like to provide materials:

How would you like to provide source materials?

  [A] Upload Files
      txt / epub novel files
      Comment export files (Qidian/Jinjiang/Fanqie)
      Weibo/WeChat article screenshots or exports

  [B] Manual Import
      Provide novel file path
      Provide comment link (manual collection needed)

  [C] Paste Text
      Copy-paste text directly

Can mix and match, or skip entirely (generate from manual info only).

---

Option A: Upload Files

• PDF / Images: Read tool directly
• txt novel files:

  python3 ${CLAUDE_SKILL_DIR}/tools/novel_parser.py --file {path} --output /tmp/novel_out.txt
  

  Then Read /tmp/novel_out.txt
• epub novel files:

  python3 ${CLAUDE_SKILL_DIR}/tools/epub_parser.py --file {path} --output /tmp/epub_out.txt
  

  Then Read /tmp/epub_out.txt
• Comment export files:

  python3 ${CLAUDE_SKILL_DIR}/tools/comment_parser.py --file {path} --platform {qidian/jinjiang/fanqie} --output /tmp/comment_out.txt
  

  Then Read /tmp/comment_out.txt
• Weibo export:

  python3 ${CLAUDE_SKILL_DIR}/tools/weibo_collector.py --file {path} --output /tmp/weibo_out.txt
  

  Then Read /tmp/weibo_out.txt
• WeChat article export:

  python3 ${CLAUDE_SKILL_DIR}/tools/wechat_parser.py --file {path} --output /tmp/wechat_out.txt
  

  Then Read /tmp/wechat_out.txt
• Markdown / TXT: Read tool directly

---

Option B: Manual Import

When user provides file paths, use the corresponding parsing tools to read.

---

Option C: Paste Text

User-pasted content is used directly as text material. No tools needed.

---

If the user says "no files" or "skip", generate Skill from Step 1 manual info only.

Step 3: Analyze Source Material

Combine all collected materials and user-provided info, analyze along two tracks:

Track A (Writing Skill):

• Refer to ${CLAUDE_SKILL_DIR}/prompts/writing_analyzer.md for extraction dimensions
• Extract: narrative style, plot construction, character creation, world-building, update habits
• Emphasize different aspects by genre type (fantasy/urban/scifi/history/infinite)

Track B (Author Persona):

• Refer to ${CLAUDE_SKILL_DIR}/prompts/author_persona_analyzer.md for extraction dimensions
• Translate user-provided tags into concrete behavior rules (see tag translation table)
• Extract from materials: communication style, creative philosophy, interaction behavior

Step 4: Generate and Preview

Use ${CLAUDE_SKILL_DIR}/prompts/writing_builder.md to generate Writing Skill content. Use ${CLAUDE_SKILL_DIR}/prompts/author_persona_builder.md to generate Author Persona content (5-layer structure).

Show the user a summary (5-8 lines each), ask:

Writing Skill Summary:
  - Narrative style: {xxx}
  - Plot construction: {xxx}
  - Character creation: {xxx}
  ...

Author Persona Summary:
  - Core personality: {xxx}
  - Communication style: {xxx}
  - Interaction behavior: {xxx}
  ...

Confirm generation? Or need adjustments?

Step 5: Write Files

After user confirmation, execute the following:

1. Create directory structure (Bash):

mkdir -p authors/{slug}/versions
mkdir -p authors/{slug}/knowledge/novels
mkdir -p authors/{slug}/knowledge/comments
mkdir -p authors/{slug}/knowledge/social

2. Write writing.md (Write tool): Path: authors/{slug}/writing.md

3. Write author_persona.md (Write tool): Path: authors/{slug}/author_persona.md

4. Write meta.json (Write tool): Path: authors/{slug}/meta.json Content:

{
  "name": "{pen_name}",
  "slug": "{slug}",
  "created_at": "{ISO_timestamp}",
  "updated_at": "{ISO_timestamp}",
  "version": "v1",
  "profile": {
    "platform": "{platform}",
    "level": "{level}",
    "masterpiece": "{masterpiece}",
    "gender": "{gender}"
  },
  "tags": {
    "writing_style": [...],
    "update_habit": [...],
    "personality": [...],
    "genre": [...]
  },
  "impression": "{creative_impression}",
  "knowledge_sources": [...imported file list],
  "corrections_count": 0
}

5. Generate full SKILL.md (Write tool): Path: authors/{slug}/SKILL.md

SKILL.md structure:

---
name: author-{slug}
description: {pen_name}, {platform} {level} {masterpiece}
user-invocable: true
---

# {pen_name}

{platform} {level} {masterpiece}{append gender if available}

---

## PART A: Writing Capabilities

{full writing.md content}

---

## PART B: Author Persona

{full author_persona.md content}

---

## Execution Rules

1. PART B decides first: what attitude to take on this task?
2. PART A executes: use your writing skills to complete the task
3. Always maintain PART B's communication style in output
4. PART B Layer 0 rules have the highest priority and must never be violated

Inform user:

Author Skill created!

Location: authors/{slug}/
Commands: /{slug} (full version)
          /{slug}-writing (writing capabilities only)
          /{slug}-persona (author persona only)

If something feels off, just say "he wouldn't do that" and I'll update it.

---

Evolution Mode: Append Files

When user provides new files or text:

1. Read new content using Step 2 methods
2. Read existing authors/{slug}/writing.md and author_persona.md
3. Refer to ${CLAUDE_SKILL_DIR}/prompts/merger.md for incremental analysis
4. Archive current version (Bash):

   python3 ${CLAUDE_SKILL_DIR}/tools/version_manager.py --action backup --slug {slug} --base-dir ./authors
   

5. Use Edit tool to append incremental content to relevant files
6. Regenerate SKILL.md (merge latest writing.md + author_persona.md)
7. Update meta.json version and updated_at

---

Evolution Mode: Conversation Correction

When user expresses "that's wrong" / "he should be":

1. Refer to ${CLAUDE_SKILL_DIR}/prompts/correction_handler.md to identify correction content
2. Determine if it belongs to Writing (style/plot) or Persona (personality/interaction)
3. Generate correction record
4. Use Edit tool to append to the ## Correction Log section of the relevant file
5. Regenerate SKILL.md

---

Management Commands

/list-authors:

python3 ${CLAUDE_SKILL_DIR}/tools/skill_writer.py --action list --base-dir ./authors

/author-rollback {slug} {version}:

python3 ${CLAUDE_SKILL_DIR}/tools/version_manager.py --action rollback --slug {slug} --version {version} --base-dir ./authors

/delete-author {slug}: After confirmation:

rm -rf authors/{slug}

---

Tag Translation Table

The tag translation table is maintained separately at ${CLAUDE_SKILL_DIR}/tag_translations.md.

This file contains mappings from the following four tag categories to Layer 0 behavior rules:

• Writing style tags (water master, plot-tight, slow-burn, power-fantasy, angst, farming, OP-MC, mortal-MC)
• Update habit tags (daily updater, weekly updater, monthly updater, hiatus-prone, burst writer, steady)
• Personality tags (sharp-tongued, fan-pamperer, aloof, self-deprecating, jokester, serious, tsundere, zen)
• Genre tags (fantasy, urban, scifi, historical, infinite, xianxia, game, horror)

During Step 3 analysis, read this file to translate user-provided tags into concrete behavior rules.