RSS & Atom Feed Aggregator - Merge, Filter, Deduplicate

Read dozens of RSS, Atom and podcast feeds in one run and get a single, clean, deduplicated dataset you can export to CSV/Excel/JSON or send straight into Google Sheets, Slack, Make, Zapier, n8n or your AI / RAG pipeline.

Don't know the feed URL? Paste the website URL and the Actor finds the feed for you (autodiscovery).

What does RSS & Atom Feed Aggregator do?

• 📰 Reads RSS 0.9x/1.0/2.0, Atom and podcast feeds (robust parser, handles broken feeds)
• 🔍 Feed autodiscovery from normal website URLs (<link rel="alternate">)
• 🧹 Cross-feed deduplication: the same story syndicated in several feeds appears once (canonical URL without utm_* tracking, then GUID)
• 🎯 Keyword include / exclude filters (whole words, case-insensitive) and max age in days
• ✂️ HTML → plain-text summaries with a length limit, optional full content text for LLMs
• 🖼️ Images and enclosures (podcast MP3 URLs, sizes)
• 📊 Per-feed health report (which feeds failed and why) in the key-value store
• ⏱️ Schedule it hourly or daily for news monitoring and digest newsletters

Use cases

• Media monitoring: track a topic across 50 news sites, only the items mentioning your brand or keywords
• Newsletter curation: a daily table of new articles from your favourite blogs
• AI / RAG ingestion: clean text + metadata from trusted sources, ready to embed
• Podcast catalogues: episode titles, dates and audio URLs from many shows
• Competitor watch: new posts on competitors' blogs and changelogs

How much does it cost?

Pay-per-event: $0.80 per 1,000 items in your output ($0.0008 each). Items removed by your filters or by deduplication are not charged, and neither are failed feeds. Use Maximum cost per run (or Max items in total) to cap your spend.

Input example

{
  "feedUrls": ["https://hnrss.org/frontpage", "https://www.theverge.com"],
  "autodiscoverFeeds": true,
  "includeKeywords": ["AI", "open source"],
  "excludeKeywords": ["sponsored"],
  "maxAgeDays": 2,
  "maxItemsPerFeed": 0,
  "maxItems": 0,
  "deduplicate": true,
  "summaryMaxLength": 500,
  "includeFullContent": false,
  "maxConcurrency": 5
}

Output example

{
  "itemId": "3f1c2a9b0d7e4c11",
  "title": "Example headline",
  "link": "https://www.example.com/story?utm_source=rss",
  "canonicalUrl": "https://www.example.com/story",
  "guid": "https://www.example.com/story",
  "published": "2026-10-08T03:12:00+00:00",
  "author": "Newsroom",
  "categories": ["Technology"],
  "summary": "Plain-text summary trimmed to your limit…",
  "contentText": null,
  "imageUrl": "https://www.example.com/thumb.jpg",
  "enclosures": [],
  "matchedKeywords": ["AI"],
  "feedUrl": "https://www.example.com/rss.xml",
  "feedTitle": "Example News",
  "feedLink": "https://www.example.com",
  "feedLanguage": "en",
  "fetchedAt": "2026-10-08T04:00:00+00:00"
}

Also saved in the key-value store: SUMMARY (counts) and FEEDS_REPORT (status of every feed: OK, HTTP error, not a feed, no feed found on the page).

Responsible use

The Actor downloads each feed URL you provide once per run, exactly like a feed reader, with an identifying User-Agent. It does not crawl websites or open article pages (autodiscovery reads only the single page you pasted). Feed content belongs to its publishers: respect their terms and copyright when you republish it - linking to the original article is always the safe choice.

FAQ

• JSON Feed support? Not yet - tell us in Issues if you need it.
• Can it read password-protected feeds? No, public feeds only.
• Something broke? Open an issue in the Issues tab; we reply within 24 hours.

Made by Gufobyte - small, honest data tools.