News Monitor: Keyword News Search + RSS/Atom Feeds

Track any topic, brand or competitor across world news, plus any RSS or Atom feed, in one run. Give it keywords and/or feed URLs and get a clean, de-duplicated list of articles with the direct publisher link, headline, outlet, date, language and country. Turn on Only new articles since last run, schedule it, and you have a cheap news alert that only bills you for new stories.

Keyword search uses the open GDELT news index (100,000+ outlets in 100+ languages, updated every 15 minutes), which is free for commercial use. Feeds are read straight from the publisher. No login, no browser, no proxies.

Great for:

• 📰 Brand and competitor monitoring: who's writing about you this week
• 📈 PR, comms and investor relations: daily coverage reports
• 🤖 AI agents, newsletters and blogs: a reliable feed of fresh stories to summarise
• 🔔 Alerts: schedule hourly or daily with only-new mode, then send to Slack, email or a webhook with Apify integrations

Input

Field	What to put in	Default
queries	Keywords, one per line. "exact phrase", (a OR b), -exclude, domain:abc.net.au	–
timespan	Last hour … last 3 months	last 24 hours
language / country	e.g. english, australia	any
maxArticlesPerQuery	Newest first, up to 250	100
feedUrls	RSS/Atom feed URLs, or a website URL that advertises a feed (we find it)	–
feedKeywords	Only keep feed items mentioning one of these	keep all
onlyNewSinceLastRun	Remember returned articles and output only new ones next time	off

{
  "queries": ["\"interest rates\" RBA", "(Canva OR Atlassian) -sport"],
  "timespan": "24h",
  "country": "australia",
  "feedUrls": ["https://www.abc.net.au/news/feed/51120/rss.xml"],
  "onlyNewSinceLastRun": true
}

Output

One row per article: title, url (publisher link), domain, source, publishedAt, snippet (feeds), author, imageUrl, language, sourceCountry, categories, origin (gdelt or feed), query / feedUrl. Articles found by several queries or feeds appear once (tracking parameters are ignored when matching). A SUMMARY record lists each query and feed with how many articles were found and returned, and any errors.

Pricing

Pay per result: you're charged per article returned. Duplicates, filtered-out items, failed feeds and the SUMMARY are free. Set a maximum cost per run and the Actor stops when it's reached.

Notes and limits

• News search returns headlines and links, not article text. GDELT gives the headline, link, outlet, language, country and first-seen time; feeds add a summary. Read the full story on the publisher's site.
• Queries run one at a time with a short pause (GDELT asks for at most one request every 5 seconds), so 10 queries take about a minute.
• Very short or very common single words may be rejected by the index; the reason is shown in SUMMARY.
• Feeds must be public. Pages that aren't feeds and don't advertise one are reported in SUMMARY.

Responsible use

Articles belong to their publishers. This Actor returns headlines, short summaries and links so you can find and read stories at the source; don't republish full articles without permission. Google News is deliberately not used, because its feed terms allow personal, non-commercial use only.

Built by Eddie Power.