ai-news-digest
Skill3 starsChanged 5 months agoMITView on GitHub
Fetch and summarize the latest AI news from reputable sources (via RSS/Atom feeds) and produce a dated digest with links. Use when the user asks for “latest AI news”, “AI 新闻汇总/简报”, “本周/今日 AI 动态”, or when you need to gather up-to-date AI announcements, research highlights, and industry coverage from known sites.
- Reads credentials
2 other projects use a version of this.
Adopted by
Most recently taken up by
- SidneyRadwan/claude-skillstheir own version, first seen Oct 25, 2026
- Red-empty-potato/ai-agent-skillstheir own version, first seen Aug 29, 2026
Other versions2 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 2
- Red-empty-potato/ai-agent-skills★ 0same name
- SidneyRadwan/claude-skills★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository8 other files it gives its agents.
- agent-pipeline-orchestrationSkill
- api-data-researchSkill
- grok-x-searchSkill
- markdown-platform-packSkill
- miniapp-figma-alignmentSkill
- social-content-parserSkill
- video-to-textSkill
- weijincha-content-checkSkill
The file
Show its rules and text
We split this skill into the rules it gives an agent
14 rules. About 0 are common, 9 are distinctive, and 5 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈9
Distinctive · estimate
This author's own choices. Worth a look.
5
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 28
Use web.run.search_query with references/x_watchlist.json and references/non_rss_sources.json.
- Very specific: line 29
Keep only items within the same time window (e.g.
- Very specific: line 44
If you only need “hot list”, run faster with --no-feedback (and/or --no-profile).
- Distinctive: line 49
Always include specific dates (e.g.
- Distinctive: line 51
If items conflict across sources, mention uncertainty and cite multiple links.
- Distinctive: line 54
If the user wants only certain outlets/topics
- Distinctive: line 57
Use filters when you only need a subset:
- Distinctive: line 68
If there are many items within 24h, cluster by theme and summarize each cluster.
- Distinctive: line 72
Add 1–2 sentence summaries per item, grouped by section
- Distinctive: line 77
Prefer RSS/Atom feeds over scraping HTML for reliability.
- Very specific: line 78
If a target site has no public feed, use web.run (search + open) as a fallback and still include dated links.
- Distinctive: line 82
If a feed is down, retry once, then skip and report which sources failed.
- Distinctive: line 83
Do not claim anything is “latest” without a date/time window (e.g.
- Very specific: line 87
Use web.run.search_query with recency (days) and queries like:
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
AI News Digest
Workflow (required)
强制要求:在生成 AI 新闻汇总/简报前,必须先运行脚本获取最新数据。不要凭记忆或直接编写新闻摘要。若脚本失败,先重试一次;仍失败再使用 web.run 作为兜底,并在结果中标注“数据来源与时间窗口”。
- 必须抓取最新新闻
默认抓取过去 24 小时;如果用户指定“本周”“过去 72 小时”“今天”等范围,调整 --since。Use a task-local output path such as ./ai-news.json, ./ai-news.md, or a file under a temporary working directory. From the skill directory, run:
python3 scripts/fetch_ai_news.py --since 24h --limit 0 --max-per-source 0 --format json --output ./ai-news.jsonOr generate Markdown directly (mandatory alternative):
python3 scripts/fetch_ai_news.py --since 24h --limit 0 --max-per-source 0 --format md --output ./ai-news.md1b) 需要时补充 X + 非 RSS 站点
- Use
web.run.search_querywithreferences/x_watchlist.jsonandreferences/non_rss_sources.json. - Keep only items within the same time window (e.g. past 72 hours), and prefer links to the canonical announcement/blog/paper.
1c) 需要时补充 GitHub “热门上榜项目” + 社区热度/反馈
Run:
python3 scripts/fetch_github_ai_trending.py --since daily --limit 0 --format md --output ./github-ai.mdNotes:
- This uses GitHub Trending (HTML) for the “up榜热门” list, then enriches repos via GitHub API for heat/feedback signals.
- Set
GITHUB_TOKEN(orGH_TOKEN) to avoid low rate limits. - Feedback signals include: open PRs, open/closed issues in the past N days, “good first issue/help wanted”, and top discussed issues.
- If you only need “hot list”, run faster with
--no-feedback(and/or--no-profile).
- 生成中文汇总(必须)
- Output in Chinese unless the user asked for English.
- Always include specific dates (e.g. “2026-02-04”) for “today / latest / yesterday”.
- For each item: 发生了什么 + 为什么重要 + link(必要时 2 links).
- If items conflict across sources, mention uncertainty and cite multiple links.
- End with a Chinese wrap-up section: “趋势与建议” (3–7 bullets). For 24h digests, title it “24小时趋势与建议”.
- If the user wants only certain outlets/topics
- Edit
references/sources.json(add/remove feeds, or toggle"enabled": true/false), then re-run the script. - Use filters when you only need a subset:
- Keywords:
--keywords agent,benchmark,open-source - Source tags:
--include-tags vendor,research/--exclude-tags community
- Keywords:
Output templates
Short digest (default)
- Title:
AI News Digest — <DATE RANGE> - Sections (optional): Product/Company, Research, Policy/Safety, Market/Business
- 8–15 bullets total, each with 1 link (2 links if needed)
- If there are many items within 24h, cluster by theme and summarize each cluster; do not drop items silently.
Long digest
- Add 1–2 sentence summaries per item, grouped by section
- Include a “Notable trends” section at the end (3–5 bullets)
Notes
- Prefer RSS/Atom feeds over scraping HTML for reliability.
- If a target site has no public feed, use
web.run(search + open) as a fallback and still include dated links. - For non-RSS “known sites”, use
references/non_rss_sources.jsonquery templates withweb.run.search_query. - For X (Twitter) news, prefer
web.runsearch (no scraping): usereferences/x_watchlist.jsonhandles + templates, and set a clear time window (e.g. past 72 hours). - For GitHub trending, prefer
scripts/fetch_github_ai_trending.py(no scraping beyond the Trending listing page). - If a feed is down, retry once, then skip and report which sources failed.
- Do not claim anything is “latest” without a date/time window (e.g. “past 72 hours”).
X (Twitter) quick recipe (web.run)
Use web.run.search_query with recency (days) and queries like:
site:x.com/OpenAI (announce OR released OR launch OR model OR paper)site:x.com (OpenAI OR AnthropicAI OR GoogleDeepMind OR MistralAI OR xai) (released OR launch OR model)
Then: extract only posts that link to an official announcement/blog/paper, and include both the X link and the canonical source link when possible.
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.

