RSS, Atom & JSON Feed Reader

Turn feed URLs or ordinary website URLs into clean structured feed items. The Actor reads RSS 0.9x/2.0, RSS 1.0/RDF, Atom 1.0 and JSON Feed 1.x. For website URLs it first discovers feeds from <link rel="alternate"> tags and then tries common feed paths such as /feed, /rss.xml, /atom.xml, /feed.xml and /index.xml.

It uses plain HTTP requests only, never a browser, and it skips failed feeds without charging for them.

Use cases

• Monitor news, changelogs, blogs and podcasts from one API call.
• Convert mixed RSS, Atom and JSON Feed sources to a single dataset shape.
• Feed AI agents with recent articles filtered by sinceDate.
• Discover a site's feed URL before ingesting its entries.
• Extract podcast media fields, enclosures, images, authors and categories in bulk.

Input

Field	Default	Description
urls	–	Feed URLs or website URLs, one per line. Plain domains are accepted.
maxItemsPerFeed	50	Maximum number of entries returned from each resolved feed.
sinceDate	–	Optional ISO date. Only entries published after this date are returned.
includeContent	false	Include full HTML/content fields such as content:encoded.
stripHtml	true	Strip HTML into text. Summaries are text; when content is included, contentText is added.
maxConcurrency	10	Number of input URLs processed in parallel.
proxyConfiguration	off	Optional Apify Proxy configuration. Most public feeds work without a proxy.

{
  "urls": [
    "https://hnrss.org/frontpage",
    "https://www.theverge.com",
    "https://github.blog/feed/",
    "https://www.jsonfeed.org/feed.json"
  ],
  "maxItemsPerFeed": 5,
  "sinceDate": "2026-01-01T00:00:00Z"
}

Output

One dataset item is produced for each feed entry. The Actor also stores a run summary in the default key-value store under SUMMARY with resolved feeds, item counts and per-feed errors.

{
  "feedUrl": "https://github.blog/feed/",
  "feedTitle": "The GitHub Blog",
  "feedLink": "https://github.blog/",
  "title": "ReviewBench: An open benchmark for AI code review",
  "link": "https://github.blog/ai-and-ml/github-copilot/reviewbench-an-open-benchmark-for-ai-code-review/",
  "guid": "https://github.blog/?p=99258",
  "author": "Michelle Zhou",
  "publishedAt": "2026-10-05T15:59:40.000Z",
  "updatedAt": null,
  "summary": "We’re launching ReviewBench, a benchmark for code review agents built on representative GitHub pull requests, multi-source ground truth, calibrated evaluation, and production-aligned metrics. The post ReviewBench: An open benchmark for AI code review appeared first on The GitHub Blog.",
  "categories": ["AI & ML", "GitHub Copilot", "AI", "benchmarking", "code review"],
  "image": null,
  "enclosures": [],
  "duration": null,
  "episode": null,
  "language": "en-US"
}

Fields include feed metadata (feedUrl, feedTitle, feedLink, language), item metadata (title, link, guid, author, publishedAt, updatedAt), normalized text summary, optional full content, categories, best image, enclosures (url, type, length) and podcast fields (duration, episode).

Pricing

Pay per event: you pay only for successfully pushed feed items.

Event	Price	When
Item (feed-item)	$0.0003 ($0.30 per 1,000 items)	Each feed entry written to the dataset. Failed feeds are free.
Actor start	$0.00005	Charged when the Actor starts, according to Actor memory.

The Actor respects your maximum cost per run. When the item budget is reached, it stops cleanly and does not push uncharged items.

FAQ

Can I pass website URLs instead of feed URLs?
Yes. The Actor discovers RSS, Atom and JSON Feed links from HTML and common feed paths.

Which feed formats are supported?
RSS 0.9x/2.0, RSS 1.0/RDF, Atom 1.0 and JSON Feed 1.x, including common extensions such as dc:creator, content:encoded, media:content, media:thumbnail, enclosure and itunes:*.

Are failed feeds charged?
No. DNS errors, HTTP errors and unparseable feeds are listed in SUMMARY but are not pushed as paid dataset items.

Does it run JavaScript?
No. It performs polite HTTP-only requests with a 20 MB response cap per feed/page.

Changelog

• 0.1: Initial release. Supports RSS, RDF, Atom, JSON Feed, HTML autodiscovery, date filtering, deduplication, media extraction and pay-per-item charging.