news

name	news
description	Fetch news headlines and synthesize structured briefings from multiple sources.
requires_bins	curl
requires_env	GEMINI_API_KEY

News Digest

Aggregate news from RSS feeds and APIs, then synthesize structured briefings.

Trigger Phrases

• "今天有什么新闻"
• "科技新闻简报"
• "daily briefing"
• "news digest"
• " headlines"
• "今日要闻"

Workflow

1. Fetch headlines via web_fetch or curl
2. Rank by recency and source credibility
3. Deep-read top articles via web_fetch
4. Synthesize structured digest with write_file

RSS Sources

Category	URL
Tech	https://news.google.com/rss/topics/CAAqJggKIiBDQkFTRWdvSUwyMHZNRGRqTVhZU0FtVnVHZ0pWVXlnQVAB?hl=en-US
World	https://news.google.com/rss/topics/CAAqJggKIiBDQkFTRWdvSUwyMHZNRFZxYUdjU0FtVnVHZ0pWVXlnQVAB?hl=en-US
HN	https://hacker-news.firebaseio.com/v0/topstories.json

Example Usage

# Fetch Google News tech feed
curl -s "https://news.google.com/rss/topics/CAAqJggKIiBDQkFTRWdvSUwyMHZNRGRqTVhZU0FtVnVHZ0pWVXlnQVAB?hl=en-US"

# Fetch HN top stories
curl -s "https://hacker-news.firebaseio.com/v0/topstories.json" | head -10

Output Format

Save to news-digest-YYYY-MM-DD.md:

# News Digest — 2026-03-11

## Technology
- **Headline** — Summary [Source](url)
- **Headline 2** — Summary [Source](url)

Key takeaway: ...

## World
...

Deep Research Integration

For deeper analysis on a story, use web_search then read_file on results.

Tips

• Google News RSS returns XML; use grep or sed to extract <title> and <link>
• HN API returns story IDs; fetch details at v0/item/{id}.json
• Cache results to avoid rate limits