← 返回 Skills 市场
utopiabenben

Ai Content Tailor

作者 utopiabenben · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ✓ 安全检测通过
230
总下载
0
收藏
1
当前安装
1
版本数
在 OpenClaw 中安装
/install ai-content-tailor
功能描述
快速将文章智能改写成公众号、小红书、知乎、抖音四个平台适配版本,保持核心观点并符合平台风格要求。
使用说明 (SKILL.md)

Content Repurposer - 内容多平台适配器

一键将同一篇文章改写成适合不同平台的版本,节省内容创作者的时间。

功能

  • 📱 多平台适配:公众号、小红书、知乎、抖音四种格式
  • 🤖 AI 智能改写:使用 LLM 调整语气、长度、结构
  • 快速处理:几秒钟生成4个版本
  • 📝 保持核心:保留原文核心观点,只调整呈现方式
  • 🎯 平台优化:符合各平台特性(标题、段落、标签等)
  • 👀 预览功能:预览所有平台版本后再输出

快速开始

1. 安装

clawhub install ai-content-tailor
cd ~/.openclaw/workspace/skills/ai-content-tailor
./install.sh

2. 配置 OpenAI API

export OPENAI_API_KEY="your-api-key"

3. 使用

# 快速改写
ai-content-tailor repurpose article.md --output ./versions/

# 指定平台
ai-content-tailor repurpose article.md --platforms wechat,xiaohongshu,zhihu,dy

# 预览模式
ai-content-tailor repurpose article.md --preview

# 批量处理
ai-content-tailor batch ./articles/ --output ./repurposed/

参数

参数 类型 必需 描述
input 路径 输入文章(Markdown/Text)
command 选项 repurposebatch
--output 路径 输出目录(默认:当前目录)
--platforms 列表 平台列表:wechat,xhs,zhihu,dy(默认:全部)
--model 字符串 LLM 模型:gpt-4o/gpt-4o-mini(默认:gpt-4o-mini)
--preview 布尔 只预览,不保存文件
--tone 选项 语气:professional,casual,storytelling(默认:自动)

平台特性

公众号 (wechat)

  • 标题:简洁有力,~20字以内
  • 结构:段落清晰,每段2-3行
  • 风格:正式+亲和力
  • 配图提示:[图片] 标注

小红书 (xhs)

  • 标题:爆款风格,表情符号,~10-15字
  • 开头:引起兴趣/痛点/悬念
  • 标签:添加3-5个相关话题标签
  • 风格:口语化,真实分享

知乎 (zhihu)

  • 标题:具体/疑问/干货
  • 结构:观点+论证+案例
  • 风格:理性,专业,有深度
  • 结尾:开放式问题或总结

抖音 (dy)

  • 标题:简短有力,引发好奇
  • 结构:快速切入,分点简短
  • 风格:口语,节奏快
  • 适合口播文案

输入文章要求

  • 格式:Markdown 或纯文本
  • 长度:建议 500-3000 字
  • 内容:完整观点 + 支撑材料
  • 标题:清晰的主题

输出文件

article.md
├── wechat_article.md
├── xhs_post.md
├── zhihu_answer.md
└── dy_script.md

使用示例

# 改写一篇公众号文章为四个平台版本
ai-content-tailor repurpose my_article.md --output ./all_platforms/

# 只生成小红书和抖音版本
ai-content-tailor repurpose my_article.md --platforms xhs,dy

# 用正式语气改写
ai-content-tailor repurpose my_article.md --tone professional

# 批量处理 articles 文件夹
ai-content-tailor batch ./articles/ --output ./repurposed/ --preview

技术栈

  • LLM:OpenAI GPT-4o-mini(默认)
  • 依赖:openai, python-dotenv
  • 配置:~/.openclaw/workspace/skills/ai-content-tailor/.env

License

MIT

安全使用建议
This skill appears to do exactly what it claims: it reads article files you provide and sends them to OpenAI to produce platform-specific rewrites. Before installing, consider: (1) it will transmit the full article text to OpenAI — avoid using sensitive or private content unless you accept that. (2) Your OPENAI_API_KEY is stored in a local .env file (plaintext) in the skill directory; protect that file and revoke the key if compromised. (3) The installer uses pip to install dependencies from PyPI — review packages and run installs in a controlled environment if you prefer. (4) If you want further assurance, inspect source/repurpose.py yourself (it contains the full prompt templates and API calls). Otherwise the package is internally consistent and proportionate to its stated purpose.
功能分析
Type: OpenClaw Skill Name: ai-content-tailor Version: 1.0.0 The skill bundle is a legitimate tool for repurposing content across social media platforms (WeChat, Xiaohongshu, etc.) using the OpenAI API. The installation script (install.sh) and the main logic (source/repurpose.py) are transparent, performing standard operations like dependency installation, environment variable configuration, and file I/O without any signs of malicious intent, data exfiltration, or prompt injection.
能力评估
Purpose & Capability
Name/description (multi-platform content repurposer) match the delivered artifacts: a Python CLI that reads local articles and calls OpenAI to produce platform-specific rewrites. Declared dependency (openai) and declared env var (OPENAI_API_KEY) are expected and proportional.
Instruction Scope
Runtime instructions and the Python script only read the user-supplied article files and a local .env for the OpenAI key, then send prompts to OpenAI. This is within scope, but be aware article contents (and any content in the input files) are transmitted to OpenAI — not to any other external endpoints.
Install Mechanism
install.sh is a simple script that checks for python3, pip-installs openai and python-dotenv from PyPI, creates a .env template and output directory, and marks the repurpose.py executable. No downloads from untrusted URLs or archive extraction are performed.
Credentials
Only OPENAI_API_KEY is required (and is declared in skill.json). No unrelated credentials, system config paths, or additional secrets are requested. The key is stored/loaded from a local .env file (plaintext), which is expected but carries the usual secrecy considerations.
Persistence & Privilege
Skill is not always-enabled, does not modify other skills, and only writes files under its own workspace (creates .env and output directory). Uninstall.sh offers optional deletion of those files. No elevated privileges requested.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install ai-content-tailor
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /ai-content-tailor 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.0
- Initial release of ai-content-tailor. - Instantly repurpose any article for WeChat Official Account, Xiaohongshu, Zhihu, and Douyin. - Supports automatic platform-specific tone, structure, and formatting adjustments using LLM. - Batch processing, preview mode, and selective platform output available. - Simple CLI with options for tone, model, and output directory. - Retains original article’s core ideas while optimizing for each platform’s style.
元数据
Slug ai-content-tailor
版本 1.0.0
许可证 MIT-0
累计安装 1
当前安装数 1
历史版本数 1
常见问题

Ai Content Tailor 是什么?

快速将文章智能改写成公众号、小红书、知乎、抖音四个平台适配版本,保持核心观点并符合平台风格要求。 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 230 次。

如何安装 Ai Content Tailor?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install ai-content-tailor」即可一键安装,无需额外配置。

Ai Content Tailor 是免费的吗?

是的,Ai Content Tailor 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

Ai Content Tailor 支持哪些平台?

Ai Content Tailor 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 Ai Content Tailor?

由 utopiabenben(@utopiabenben)开发并维护,当前版本 v1.0.0。

💬 留言讨论