← Back to Skills Marketplace
swordman20

AI情报飞书文档

by swordman20 · GitHub ↗ · v0.0.1
cross-platform ⚠ suspicious
338
Downloads
0
Stars
1
Active Installs
1
Versions
Install in OpenClaw
/install ai-news-feishu-doc
Description
飞书AI情报
README (SKILL.md)

AI News Digest - AI 早报生成器

自动生成 AI 行业早报,支持 RSS 订阅、智能摘要、多语言分类。

快速开始

# 生成今日早报
cd scripts
python generate-ai-news.py

# 或
python3 generate-ai-news.py

# 输出文件:AI早报_2026年2月28日.md

功能特性

  • RSS 订阅 - 自动抓取配置源的最新文章
  • 智能摘要 - 提取文章要点,生成摘要
  • 关键词过滤 - 按关键词匹配优先级排序
  • 多语言分类 - 自动区分中文/英文内容
  • 媒体预览 - 提取文章配图
  • Markdown 输出 - 结构化早报文档

配置说明

配置文件:assets/ai-news-rss.yaml

feeds:
  - name: "36氪"
    url: "https://www.36kr.com/feed"
    category: "国内AI"
    language: "zh"

filters:
  keywords:
    - "GPT"
    - "Claude"
    - "AI"
    - "大模型"
    - "OpenAI"

output:
  max_articles_per_feed: 5
  total_max_articles: 15

自定义配置

  1. 添加 RSS 源:在 feeds 列表中添加新的源
  2. 调整关键词:修改 filters.keywords 匹配感兴趣的内容
  3. 输出格式:调整 output 参数控制文章数量

定时任务

添加到 OpenClaw cron 实现每日自动推送:

{
  "name": "ai-morning-news",
  "schedule": "0 8 * * *",
  "command": "python skills/ai-news-digest/scripts/generate-ai-news.py"
}

输出示例

# 📰 AI 早报 | 2026年2月28日

## 🔥 头版头条
### OpenAI 获得 1100 亿美元新投资...

## 🇨🇳 国内 AI 动态
...

## 🌍 海外 AI 动态
...

依赖

  • Python 3.7+
  • PyYAML: pip install pyyaml

🦞 Powered by OpenClaw | v1.0.0

Usage Guidance
这套技能确实会抓取配置文件里的 RSS 源、生成摘要并写出本地 Markdown,但它并不会把内容发送到飞书——如果你期待自动推送到飞书/文档,需要额外实现相应的 Feishu API 集成并提供 Feishu 的凭据。确认你希望脚本在你的环境联网抓取 RSS(会访问外部站点和嵌入图片 URL),并审阅 assets/ai-news-rss.yaml 中的来源与关键词;如需自动推送到企业聊天/文档,要求作者或你自己补充安全的发布实现并明确声明所需的 API 令牌。
Capability Analysis
Type: OpenClaw Skill Name: ai-news-feishu-doc Version: 0.0.1 The skill is designed to generate an AI news digest by fetching RSS feeds from predefined, legitimate news sources. The `SKILL.md` provides clear, benign instructions and a standard cron task. The `scripts/generate-ai-news.py` script uses `urllib.request` for network access to these sources and `yaml.safe_load` for configuration, which are appropriate and securely implemented for its stated purpose. It parses XML/HTML content and writes a Markdown report locally within the skill's directory. There is no evidence of prompt injection, data exfiltration, malicious execution, persistence mechanisms, or other harmful behaviors.
Capability Assessment
Purpose & Capability
Skill 名称与描述(“AI情报飞书文档”、“飞书AI情报”)暗示会把摘要推送到飞书/飞书文档,但代码和 SKILL.md 仅生成本地 Markdown 文件并建议将脚本加入 cron。没有任何 Feishu API 调用、回调 URL、或需要的环境变量/令牌声明,因此功能和宣称不一致,可能误导想要自动推送到飞书的用户。
Instruction Scope
SKILL.md 的运行指令限定为在本地/agent 环境运行 Python 脚本并读取 assets/ai-news-rss.yaml。脚本只访问列出的 RSS URL 和文章中的图片链接,解析并写出本地 Markdown。没有读取系统敏感路径、凭据或将数据发到第三方 API。注意:SKILL.md 提到可“推送”/定时,但并未给出任何推送到飞书的实现或外部 endpoint。
Install Mechanism
无安装规范(instruction-only + 附带脚本),仅需 Python 3.7+ 和 PyYAML。没有从不可信 URL 下载或在安装时写入磁盘之外的行为。
Credentials
不要求任何环境变量或凭据(requires.env 为空),脚本也没有引用未声明的环境变量或密钥。给出的配置只在 assets/ai-news-rss.yaml 中。
Persistence & Privilege
flags.show: always: false(默认)。技能不会请求常驻权限或修改其他技能/系统设置;仅在运行时生成一个本地 Markdown 文件。
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install ai-news-feishu-doc
  3. After installation, invoke the skill by name or use /ai-news-feishu-doc
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v0.0.1
第一个版本
Metadata
Slug ai-news-feishu-doc
Version 0.0.1
License
All-time Installs 1
Active Installs 1
Total Versions 1
Frequently Asked Questions

What is AI情报飞书文档?

飞书AI情报. It is an AI Agent Skill for Claude Code / OpenClaw, with 338 downloads so far.

How do I install AI情报飞书文档?

Run "/install ai-news-feishu-doc" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is AI情报飞书文档 free?

Yes, AI情报飞书文档 is completely free (open-source). You can download, install and use it at no cost.

Which platforms does AI情报飞书文档 support?

AI情报飞书文档 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created AI情报飞书文档?

It is built and maintained by swordman20 (@swordman20); the current version is v0.0.1.

💬 Comments