News search
Current coverage.
Clean feed.
Search current public news with one cacheable GET request. Results are normalized article entities with titles, publishers, publication dates, summaries, and public article links when available.
Request
curl --get 'https://extractor.sh/api/news' \
--data-urlencode 'q=AI infrastructure' \
--data-urlencode 'limit=10' \
--data-urlencode 'format=json'| Parameter | Required | Description |
|---|---|---|
q | Yes | A news query of up to 200 characters. |
limit | No | An integer from 1 to 50. Default: 10. |
language | No | Canonical BCP 47 language tag. Default: en-US. |
country | No | Two-letter ISO country code. Default: US. |
timeframe | No | any, 1h, 1d, 7d, or 30d. Default: any. |
format | No | json (default) or markdown. |
JSON response
JSON uses schema version 1 with type: "feed" and source: "google-news". Each entry in items is an article. Unavailable publishers, dates, or summaries remain nullable or are omitted according to the shared schema.
{
"schemaVersion": 1,
"type": "feed",
"source": "google-news",
"id": "AI infrastructure",
"url": "https://extractor.sh/api/news?q=AI+infrastructure&limit=10&format=json",
"title": "News results for AI infrastructure",
"author": null,
"publishedAt": "2026-08-02T10:00:00.000Z",
"content": "# News results for AI infrastructure\n\n...",
"media": [],
"attributes": {
"feedType": "news-search",
"query": "AI infrastructure",
"language": "en-US",
"country": "US",
"timeframe": "any",
"resultCount": 1
},
"items": [
{
"type": "article",
"source": "google-news",
"id": "…",
"url": "https://news.google.com/…",
"title": "…",
"author": "Example Publisher",
"publishedAt": "2026-08-02T10:00:00.000Z",
"content": "Public article summary…",
"media": [],
"attributes": { "publisher": "Example Publisher" }
}
]
}Behavior
- Results contain public news metadata and summaries, not complete publisher article bodies.
- Successful searches are cached for up to one hour. Cache hits bypass extraction work and are not billed.
- Uncached requests use the standard 60 requests per client per minute limit.
- Language and country select an edition. A recent timeframe excludes undated items and articles older than its inclusive cutoff; a valid empty result is returned with HTTP 200.
- Pagination and personalized feeds are not supported.
- Result availability, order, and freshness can change as public coverage changes.