content-tracker

内容追踪技能。高时效性话题追踪、链接更新监控、第三方热点聚合。 USE FOR: - "追踪AI行业动态", "追踪某某话题最新进展" - "监控这个链接的更新", "这个网站有新内容吗" - "今天的热点", "最新新闻" - 定期简报生成（配合日程使用）

Security Scan

The skill's requirements and instructions are generally consistent with a content-tracking tool; nothing requests unrelated credentials or suspicious installs, though there are small documentation mismatches to check before use.

Purpose & Capability

The name/description (topic/link/news tracking and periodic briefs) matches the declared runtime expectations (uses Python3, saves Markdown reports, depends on a separate topic_tracking skill). Required binaries (python3) are appropriate for the stated purpose. The declared dependency on the topic_tracking skill is reasonable for modular design.

Instruction Scope

SKILL.md limits behavior to fetching/filtering third-party content, generating summaries, and writing Markdown reports under ./我的追踪/{date}/. It explicitly excludes login-only content and JS-heavy pages. It does not instruct reading unrelated system files or secret env vars.

Install Mechanism

There is no install spec (instruction-only), so nothing is downloaded or written during install. This is the lowest-risk model and consistent with the skill's description.

Credentials

No credentials or config paths are requested, which is appropriate. Minor inconsistencies: the metadata lists bash and python3 as required binaries but the doc only describes Python + requests (no explicit use of bash explained). The SKILL.md mentions the Python 'requests' library but does not declare how it will be provided — the skill assumes the runtime already has that dependency. These are documentation/packaging gaps rather than direct security red flags.

Persistence & Privilege

always is false and the skill does not request system-wide changes or other skills' configs. It writes outputs to a local relative path (./我的追踪/...), which is normal for report generation.

Assessment

This skill appears coherent for tracking topics and saving Markdown reports, but verify a few practical points before installing: 1) Ensure the runtime has Python3 and the 'requests' package (the SKILL.md expects requests but there is no install step). 2) Confirm why 'bash' is listed as a required binary (the docs only describe Python usage); if you run in a constrained environment, test the skill in an isolated workspace first. 3) The skill depends on another skill 'topic_tracking'—inspect that skill's code/permissions before enabling. 4) Outputs are written to ./我的追踪/... on the agent host—avoid running on machines with sensitive data or expose those report files unintentionally. 5) Be aware of scraping and site terms of service for sources like Weibo/36Kr/Tencent; the skill excludes login and JS-heavy pages but may still trigger rate limits or blocking. If you need stronger assurance, ask the publisher for an install spec or the topic_tracking implementation to review.

内容追踪技能

高时效性内容追踪工具，支持话题搜索、链接监控、热点聚合。

核心能力

1. 话题追踪

• 智能搜索接口，过滤高时效性内容
• 按事件主体去重，避免重复推送
• 支持首次/非首次执行模式

2. 链接追踪

• 监控指定URL的内容更新
• 自动识别新增内容
• 生成更新简报

3. 第三方数据源

• 微博热搜 (weibo_hot)
• 36氪 (thirty_six_kr)
• 腾讯新闻 (tencent_news)
• 华尔街见闻 (wallstreet_cn)
• 每日热点汇总 (daily_report)

使用场景

✅ 适用：

• 行业动态追踪（AI、金融、科技等）
• 竞品/新闻监控
• 定期简报生成
• 链接更新通知

❌ 不适用：

• 需要登录的内容
• 动态JS渲染的页面
• 视频为主的页面

输出格式

所有追踪结果保存为 Markdown 简报：

./我的追踪/{yyyy_mm_dd}/{话题名称}.md

简报内容：

• 来源标题（引用链接）
• 核心内容总结
• 时效性标注

快速开始

用户：追踪一下AI Agent行业的最新动态
AI：[执行话题追踪] → 生成简报 → 返回链接

配合日程使用

可创建重复日程，自动定期追踪：

场景	频率	建议
行业动态	每日	下午3点执行
热点追踪	每日	早晚各一次
链接监控	每周	周一执行

依赖

• 技能：topic_tracking（需预先安装）
• 环境：Python3 + requests

注意事项

• 首次执行使用 --is-first-time true 参数
• 无结果表示确实没有新内容，不会重复推送
• 内容来源于原文总结，不做推断或编造