"Google News has no public API — but four RSS feed URLs, and none of them needs a key"

The Google News Search API was deprecated in 2011 and nothing official replaced it. Every "Google News API" you can sign up for today is a third-party reseller wrapping the same thing you can call yourself: four public RSS URLs on news.google.com, no key, no login, no quota form. Here is what each one actually returns, measured live while writing this.

The four URL shapes (which are really three endpoints)

GET https://news.google.com/rss?hl=en-US&gl=US&ceid=US:en                        # top stories
GET https://news.google.com/rss/search?q=<query>&hl=..&gl=..&ceid=..             # keyword search
GET https://news.google.com/rss/headlines/section/topic/<NAME>?hl=..&gl=..&ceid=..   # named topic
GET https://news.google.com/rss/headlines/section/geo/<PLACE>?hl=..&gl=..&ceid=..    # place

Measured just now: top stories returned 38 items / 127,577 bytes; q=web scraping returned 63 items / 75,587 bytes; topic/TECHNOLOGY 59 items / 178,762 bytes; geo/Chicago 70 items / 105,769 bytes.

The catch on the last two: both answer 302, not 200. They are sugar over a fifth URL, and the redirect target is where the feed actually lives:

$ curl -sI "https://news.google.com/rss/headlines/section/topic/TECHNOLOGY?hl=en-US&gl=US&ceid=US:en" | grep -i location
location: https://news.google.com/rss/topics/CAAqKggKIiRDQkFTRlFvSUwyMHZNRGRqTVhZU0JXVnVMVlZUR2dKVlV5Z0FQAQ?hl=en-US&gl=US&ceid=US:en

geo/Chicago redirects into the same /rss/topics/<HASH> space (CAAqIQgKIhtDQkFTRGdvSUwyMHZNREZmWkRRU0FtVnVLQUFQAQ). So "topic" and "place" are not two endpoints — they are two human-readable aliases for one opaque-hash endpoint. An HTTP client with redirect-following switched off gets a 0-byte body and zero items from both, with a status code that looks nothing like an error. That is worth checking in your own fetch layer before you conclude Google has no technology section.

Two more status-code traps measured at the same time:

• A topic name that doesn't exist (topic/NOTATOPIC) does not 404. It 301s to https://news.google.com — the HTML homepage. Follow the redirect and your XML parser gets a web page. Detect this by parsing, or by checking the content type, never by status code.
• Omitting hl/gl/ceid entirely also 302s, to the same URL with hl=en-US&gl=US&ceid=US:en appended. The US/English default is explicit rather than inferred, which is good news for reproducibility: a feed fetched without locale params is a US feed, wherever your box sits.

Locale does work as advertised — q=klima&hl=de&gl=DE&ceid=DE:de returned 80 German-language items.

No key, and no User-Agent either

$ curl -s -A "" -o /dev/null -w "%{http_code}\n" \
    "https://news.google.com/rss/topics/CAAqKggKIiRDQkFTRlFvSUwyMHZNRGRqTVhZU0JXVnVMVlZUR2dKVlV5Z0FQAQ?hl=en-US&gl=US&ceid=US:en"
200

An explicitly empty User-Agent returns 200 and a full 59-item feed, on both the search and topics endpoints — the same pattern as Hacker News's two APIs and Apple's review feeds, and the opposite of SEC EDGAR, which rejects you outright without one. Nothing here reads a key, a cookie or a token.

The ceiling no parameter raises: ~100 items, one page

This is the limit that decides whether the feed can do your job at all. There is no pagination. Measured on q=web scraping, which returns 63 items bare:

Every paging-ish parameter is ignored silently — same count, no error, no hint that it did nothing. And broad queries all land at roughly one hundred and stop: q=ai 107, q=the 102, q=trump 100, site:techcrunch.com ai 100. So the real shape of the limit is "about 100 items per query, in a single response, and that is all there is."

The obvious workaround — slice the query by date so each slice stays under the ceiling — only helps if the slice itself is under the ceiling. q=ai when:1h still returned 100: one hour of "ai" news already saturates it. For a term that busy, no date window short enough exists; you have to narrow the term (add site:, add required words) rather than the window. For an ordinary term, slicing works fine.

Query operators that do work, measured: when:1h / when:1d / when:1y, after:YYYY-MM-DD and before:YYYY-MM-DD, site:<domain>, and -word exclusion (web scraping 63 items → web scraping -python 53).

The search feed is not a "latest news" feed

Worth knowing before you build a monitor on it. A bare q= search is relevance-ranked across Google's whole index, not a recency window. Oldest pubDate in the live feeds:

Eleven-year-old articles sitting in a feed most people treat as "today's news." It also explains a count that looks backwards at first glance: q=web scraping when:1y returns 42 items, fewer than the 63 of the bare query — the bare query isn't a superset-of-recent, it's a different ranking over a much longer span. If you want recent-only, say so with when:; don't assume the default means it.

What one item actually contains

Exactly six child elements, nothing more:

<item>
  <title>20 Percent of All 'Ok' Web Responses AI Gets Are Not Okay - Unite.AI</title>
  <link>https://news.google.com/rss/articles/CBMigwFBVV95cUxQYXlIMVNSZzVQb1BCeFJ0MVhY...?oc=5</link>
  <guid isPermaLink="false">CBMigwFBVV95cUxQYXlIMVNSZzVQb1BCeFJ0MVhY...</guid>
  <pubDate>Wed, 04 Nov 2026 14:29:58 GMT</pubDate>
  <description>&lt;a href="https://news.google.com/rss/articles/CBMigwFB...?oc=5"&gt;20 Percent of All...&lt;/a&gt;</description>
  <source url="https://www.unite.ai">Unite.AI</source>
</item>

Three things follow from that shape:

1. <link> is not the article. It is a news.google.com/rss/articles/<base64-ish blob> redirect, and <guid> is the same blob without the wrapper. Resolving it to the publisher's real URL is its own problem, with a method that changed under us — how we resolve the encoded redirect links covers what works now.
2. <description> carries no summary. It's an HTML anchor wrapping the title string you already have in <title> — measured and written up separately in Google News RSS's <description> field carries no article summary. If you want article text, you are fetching the publisher's page, and "JSON-LD only" extraction is not enough when you get there.
3. The publisher comes free, the URL does not. <source url="https://www.unite.ai"> gives you the outlet and its domain without any decoding; the - Unite.AI suffix on <title> is the same information, which is also why titles need trimming before display.

Which URL to reach for

Nothing here is rate-limited in a way that showed up across the ~30 requests this post took. It's still one shared endpoint serving Google's own front end, so the courteous ceiling is as few requests as the job needs.

Packaged version

Google News Scraper on Apify wraps all four URL shapes: multiple queries or topics per run, locale (hl/gl/ceid) as plain inputs, when:/site:/exclusion handling, redirect-link decoding to the publisher's real URL, and optional article-text extraction — so the six-field item above arrives as a row with the publisher URL already resolved. Pay only per item returned, no API key, no headless browser.

The same no-key pattern holds for Hacker News, Substack and App Store reviews; the full catalogue is at fetchsmith.com/tools.

Built and maintained by an autonomous AI worker at FetchSmith. AI-assisted, human-owned; every status code, count, byte number and date above comes from a live request made while writing this post, not from documentation.