RSS Feed Reader & Finder: RSS, Atom, JSON Feed and Podcasts

Read any RSS, Atom, RDF or JSON Feed and get clean, structured items. You can also paste a plain website, blog or podcast page URL. The Actor finds the feed for you through <link rel="alternate"> tags or common paths like /feed and /rss.xml.

Use it to monitor news, blogs, competitors, job boards, press releases, release notes and podcasts. You can also feed fresh content into AI agents, Slack, Google Sheets, n8n, Make or Zapier.

Why this RSS reader

• Every feed format. RSS 0.9x, 1.0 (RDF) and 2.0, Atom 1.0, and JSON Feed 1.0 and 1.1. It handles any character encoding (UTF-8, ISO-8859-1, Windows-1251 and others).
• Feed discovery. Give it theverge.com and it reads theverge.com/rss/index.xml.
• Clean output. Dates are normalized to ISO 8601 and HTML is turned into plain-text summaries. Relative links become absolute URLs, and each item gets an image (from media:content, media:thumbnail, itunes:image, enclosure or the first <img>).
• Podcast support. Each episode includes the audio enclosure URL, MIME type, size and duration in seconds.
• Filters. Use publishedAfter ("24 hours", "7 days" or 2026-01-01), keywords, a per-feed cap and a total cap. Duplicates across feeds are removed.
• Full content on request. Turn on includeContent to add contentHtml and contentText when the feed carries the full article.
• Fast and cheap. It sends plain HTTP requests with no browser, reads 8 feeds in parallel, and runs at the default 256 MB.

Pricing

Pay per event, with no subscription:

	This Actor	Typical alternatives
Per 1,000 items	$0.30	$1.15 – $10
Per run start	$0.0005	up to $0.035

Feeds that fail return no rows and cost nothing beyond the start fee. Set a spending limit and the run stops cleanly when it is reached.

Input

{
  "feedUrls": ["https://feeds.bbci.co.uk/news/world/rss.xml", "https://github.blog", "https://www.theverge.com"],
  "maxItemsPerFeed": 20,
  "publishedAfter": "7 days",
  "keywords": ["AI", "security"],
  "includeContent": false
}

Output

{
  "title": "YouTube might be testing hiding the dislike button",
  "link": "https://www.theverge.com/streaming/1015099/youtube-dislike-button-hide",
  "guid": "https://www.theverge.com/?p=1015099",
  "author": "Jay Peters",
  "publishedAt": "2026-10-21T00:00:56.000Z",
  "updatedAt": "2026-10-21T00:00:56.000Z",
  "summary": "YouTube appears to be testing hiding the dislike button on videos...",
  "categories": ["News", "Streaming", "YouTube"],
  "imageUrl": "https://platform.theverge.com/wp-content/uploads/....jpg",
  "enclosure": null,
  "durationSecs": null,
  "feedTitle": "The Verge",
  "feedUrl": "https://www.theverge.com/rss/index.xml",
  "siteUrl": "https://www.theverge.com/",
  "feedFormat": "atom",
  "feedLanguage": "en-US",
  "inputUrl": "https://www.theverge.com",
  "scrapedAt": "2026-10-21T05:19:08.794Z"
}

URLs that fail are listed in the FAILURES record of the run's key-value store, with the reason.

Monitoring feeds on a schedule

Create an Apify Schedule (for example, every hour) with publishedAfter: "1 hour". Each run then returns only new items. Connect the dataset to Slack, email, Google Sheets or a webhook.

Use it from code or AI agents

curl -X POST "https://api.apify.com/v2/acts/best_apis~rss-feed-reader/run-sync-get-dataset-items?token=YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"feedUrls":["https://news.ycombinator.com/rss"],"maxItemsPerFeed":10}'

The Actor also works as a tool through the Apify MCP server, so Claude, ChatGPT and Cursor agents can read feeds.

FAQ

Can I paste a homepage instead of a feed URL? Yes. Discovery is on by default. Turn on allDiscoveredFeeds to read every feed a page declares, such as category feeds.

How many items does a feed return? Whatever the publisher includes, usually the latest 10 to 100. Podcast feeds often carry the whole back catalogue.

Is it legal? Feeds are published so that software can read them. This Actor only reads public feeds and pages, and does not log in. Respect each publisher's terms when you reuse the content.

Something broken? Open an issue on the Issues tab. We usually reply within 24 hours.