News Scraper: Google News, Bing & Any RSS Feed

Track any topic across Google News, Bing News and any RSS or Atom feed: in one normalised table.

RSS is a publish-for-syndication format, so there's no key, no login and no bot wall anywhere in this Actor.

Three ways in, one shape out

Search: searchQueries: ["tesla recall", "\"interest rates\" ECB"] Full Google News search syntax: quotes, OR, site:.

Sections: topics: ["technology", "business"] Google's curated feeds: world nation business technology entertainment sports science health.

Any feed: feedUrls: ["https://github.blog/feed/", "https://hnrss.org/frontpage"] A publisher's own feed, a company blog, a subreddit, a podcast. RSS 2.0 and Atom both work, they disagree about nearly every field name, and this Actor reconciles them so a mixed list still returns one clean table.

Do you need an API key? No. Is there a free tier? Yes.

The two things people search for in this category, measured, are free and API key: google news api free, google news api key, google news api key free.

So, plainly:

• No API key. Nothing to register for, nothing to rotate, no key to leak in a repo. The source is a public API published deliberately by its owner.
• No proxy setup. The source does not bot-wall datacenter addresses, so the default works.
• Pay per result, not per month. There is no subscription and no minimum. Rows that error or that your filters drop are not charged.
• Free to try. Run it with the prefilled input and see real rows before deciding anything.

What you get

Field	Example
title	Apple options are doing something unusual into earnings
publisher	CNBC
publishedAt	2026-07-27T10:58:27Z, normalised to UTC across every feed format
summary	plain text, HTML stripped
url, image, categories, feedTitle, guid
source	which query or feed it came from

Syndicated duplicates removed

One wire story runs verbatim under dozens of outlets. dedupeHeadlines (on by default) keys on the normalised headline, not the URL, which differs per outlet, so those collapse into one row. Duplicates are not charged.

Filters

{
  "searchQueries": ["openai"],
  "keywords": ["funding", "lawsuit"],
  "excludeKeywords": ["opinion"],
  "publishers": ["Reuters", "Bloomberg"],
  "publishedSince": "2026-07-01",
  "maxArticlesPerSource": 100
}

Articles from feeds that omit a publish date are kept, not silently dropped, missing isn't the same as too old.

Any language, any country

language: "de", country: "DE" gives you the German edition. Works for he/IL, fr/FR, es/ES and the rest.

Who this is for

• Brand & competitor monitoring: a query per brand, scheduled hourly
• PR & comms: every mention, deduplicated, with publisher and timestamp
• Newsletters & aggregators: merge many feeds into one clean source
• Trading & research: headline flow on a ticker or sector, filtered by outlet
• AI / dataset builders: normalised article metadata from any feed on the web

Time window

timeframe limits how far back a Google News search reaches, and Google applies it at the source rather than this Actor filtering afterwards. Past hour returned 38 articles where an unbounded search returns 100, so you are not charged for articles outside the window you asked for.

Choice	Sent as
Past hour	when:1h
Past 12 hours	when:12h
Past 24 hours	when:1d
Past 3 days	when:3d
Past week	when:7d
Past 30 days	when:30d
Past year	when:1y

A week is 7d and a month is 30d on purpose. Google's operator understands hours, days and years; when:1w and when:1m are accepted without complaint and return zero articles. Measured twice on two queries. Anything offering you "1 week" or "1 month" against this endpoint is either translating it, or handing you an empty result with nothing in the log to explain it.

It applies to Google News searches only. Custom RSS feeds and Bing return whatever their feed holds, and publishedSince still filters those after fetch.

Notes

• Google News links are Google redirect URLs. Google stopped exposing the publisher's direct URL in its RSS output; the link opens the article through Google News, and the publisher field tells you the outlet. Custom feedUrls return the publisher's real URL directly, use those when you need the canonical link.
• Google News returns a headline and a short summary, not full article text.
• A feed that's down or isn't valid XML comes back as one error row and is not charged; the other sources in the run still return.
• Set a max charge per run in the run options for a hard spend ceiling. The Actor honors it and stops cleanly.

Pricing

Pay per result. One row = one article. Failed feeds come back as error rows and are not charged, and neither are duplicates or anything your filters removed.