ai-hotspot-collector
Skill2 starsChanged 8 months agoApache-2.0View on GitHub
Aggregate AI-related candidates from multiple built-in source fetchers (TechCrunch, Engadget, Fast Company, Hacker News, GitHub Trending, The Verge), save them into one review queue, and draft Chinese Markdown articles in a review-first workflow. Use when the user wants to 收集 AI 热点、做 AI 新闻选题池、批量看多个来源今天有什么 AI 内容、或基于多来源热点写中文稿。
In this repository27 other files it gives its agents.
- article-multi-publisherSkill
- article-pipeline-skillSkill
- character-ip-risk-checkSkill
- chrome-extension-devSkill
- chrome-extension-e2e-automationSkill
- chrome-extension-publishSkill
- chrome-extension-social-promoSkill
- chrome-webstore-image-generatorSkill
- code-reviewSkill
- fund-querySkill
- github-trending-writerSkill
- lark-workflow-doc-to-slidesSkill
- notionSkill
- playwright-twitter-accessSkill
- post-to-xhsSkill
- rewrite-skillSkill
- small-capital-investingSkill
- technical-article-preflightSkill
- technical-article-reviewSkill
- technical-spoken-rewriteSkill
- tencent-dev-community-publisherSkill
- toutiao-publisherSkill
- url-readerSkill
- wechat-article-formatterSkill
- wechat-article-generatorSkill
- wechat-publisherSkill
- wechat-sticker-publisherSkill
The file
Files
AI Hotspot Collector
This skill is a source aggregator, not a site-specific scraper.
It uses repo-local fetchers, merges candidates into one queue, and hands approved drafts to the local publishing chain.
Unified Entry
From the repository root:
python3 ai-hotspot-collector/scripts/run.py <command> [options]Commands:
fetch: 批量抓取多个来源并生成总 manifestpublish: 把某一篇已审稿内容交给 article-multi-publisher
What It Produces
默认输出到:
content/{YYYY-MM-DD}/典型结构:
content/{date}/
├── 00-今日AI热点速读.md
├── images/
├── manifest.json
├── all_candidates.json
├── summary.json
├── README.md
├── techcrunch/
├── the-verge/
├── hn/
├── twitter/
├── github-trending/
├── engadget/
└── fast-company/Workflow
1. 收集候选
默认抓全部来源:
python3 ai-hotspot-collector/scripts/run.py fetch限制来源和每源条数:
python3 ai-hotspot-collector/scripts/run.py fetch \
--sources hn,techcrunch,engadget \
--limit-per-source 3抓取阶段会:
- 为每个来源生成子目录
- 去重同事件候选
- 生成统一
manifest.json - 生成总稿
00-今日AI热点速读.md - 复制首图和关键配图到总稿
images/
2. 向用户展示候选
优先展示:
- 全局编号
- 标题
- 来源
- 作者与发布时间
- 原文链接
- 本地
content.md路径
如果某个来源抓取失败,也要在总统计里说明,不要把失败伪装成“没有新闻”。
3. 写稿并先给用户审阅
默认先把稿子写回候选目录,再给用户审阅。
推荐文件名:
article.mdtranslation-draft.zh.mdreview-draft.zh.md
如果是技术类文章,默认先走 technical-article-review,再把修订稿给用户看。
4. 审阅通过后再发布
python3 ai-hotspot-collector/scripts/run.py publish <content-root> \
--index 3 \
--source article.md常用参数:
python3 ai-hotspot-collector/scripts/run.py publish <content-root> \
--index 3 \
--platforms wechat,xhs,toutiao \
--dry-runQuality Bar
- 不要把聚合器当成单一来源
- 来源抓取逻辑已内置在 repo 内,不依赖外部 source skill 才能运行
- 不要把 RSS 摘要误写成完整报道
- 跨平台重复事件优先挑信息最完整的一篇做主稿
- 如果启用了 LLM 重写,也不能删除图片、链接和来源信息
References
- 文章模板:references/article-template.md
- 来源编辑提示:references/platform-notes.md
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

