RSS & Atom Feed Monitor - New Items Only + Feed Finder avatar

RSS & Atom Feed Monitor - New Items Only + Feed Finder

Pricing

from $0.56 / 1,000 new feed items

Go to Apify Store
RSS & Atom Feed Monitor - New Items Only + Feed Finder

RSS & Atom Feed Monitor - New Items Only + Feed Finder

Read any RSS, Atom, RDF or JSON Feed (or just paste a website / YouTube channel URL and it finds the feed). Schedule it and get only NEW items since the last run, with keyword and date filters. Pay only for new items.

Pricing

from $0.56 / 1,000 new feed items

Rating

0.0

(0)

Developer

Lindenbyte Studio

Lindenbyte Studio

Maintained by Community

Actor stats

0

Bookmarked

2

Total users

1

Monthly active users

4 hours ago

Last modified

Features

Limited permissions

Share

RSS & Atom Feed Monitor: New Items Only + Feed Finder

Read any RSS, Atom, RSS 1.0/RDF or JSON Feed and get clean, normalised items: title, link, date, author, plain-text summary, categories, image, and podcast/video enclosures. You don't need the feed URL: paste a website, blog, news site or YouTube channel and the Actor finds the feed for you. Then schedule it. In monitoring mode each run outputs only the items that are new since the last run, and you pay only for those new items.

Step-by-step tutorial: Get alerts for new blog posts and YouTube videos without duplicates

Why this one

  • New items only, across runs. The Actor remembers every item it has seen in each feed (by guid, link, or title + date) and outputs just the new ones. Schedule it hourly or daily for a clean stream of new posts, releases, videos or episodes, with no duplicates to clean up in Zapier, Make or Slack.
  • Pay only for new items. Items you already got, items your filters remove, and the silent baseline are all free. A feed with nothing new costs nothing beyond the small start fee.
  • Finds the feed for you. For a homepage or blog URL it reads the <link rel="alternate"> feed tags and then tries the common paths (/feed, /rss, /feed.xml, /rss.xml, /atom.xml, /index.xml, /feed.json). YouTube channel/UC..., @handle and playlist URLs become their official video feeds.
  • Feed finder mode. Paste a list of websites and get one row per working feed, with its format, item count and the date of the latest item. It also shows you which blogs are dead.
  • Handles real-world feeds. RSS 2.0, RSS 0.9x, RSS 1.0/RDF, Atom 1.0, JSON Feed 1.0/1.1, CDATA, HTML entities, Atom XHTML content, gzip, byte-order marks, non-UTF-8 encodings (e.g. ISO-8859-1), and wrong Content-Type headers.
  • Keyword and date filters. Include/exclude keywords (matched in the title and summary) and publishedAfter (2026-11-01 or relative, like 7 days).
  • Polite and fast. Conditional GET (ETag / Last-Modified), so an unchanged feed is a tiny HTTP 304. A per-host throttle, retries with backoff on 429/5xx, and a pure HTTP client: no browser, no proxy.

Use cases

  • Content monitoring. Track industry blogs, newsletters, changelogs and documentation updates.
  • Competitor blog watch. Get every new post from competitors' blogs the day it is published (pair it with our Sitemap Monitor for pages that are not in a feed).
  • News alerts to Slack, email, Zapier or Make. Schedule the Actor and connect an integration or webhook ("Run succeeded", then read the dataset). Only new items arrive, so each alert is new.
  • Podcast new-episode tracking. Get the audio enclosure URL, file size and episode artwork for every new episode.
  • YouTube upload alerts. Paste channel URLs and get every new video with its link, description and thumbnail.
  • GitHub release tracking. Use https://github.com/<owner>/<repo>/releases.atom to follow releases of the libraries you depend on.
  • Security advisories. Monitor vendor and distribution advisory feeds, with keywords for the products you run.
  • Feed finder. Find the RSS feed of hundreds of websites at once, for a reader, an aggregator or a lead list of active blogs.

Input

FieldWhat it does
startUrlsFeed URLs, or any website / blog / YouTube channel / playlist URL (the feed is found automatically)
onlyNewItemsMonitoring mode (default on): output only items not seen on previous runs, per feed
emitOnBaselineFirst run of a monitor: output the current items (on, the default) or just remember them (off)
statusRowWhenEmptyOn (default): a run with nothing new adds one free run-status row. Off: empty dataset
stateKeyName of the monitor; different names keep separate memories
maxItemsPerFeedCost cap: max items output per feed per run, newest first (default 50)
includeKeywords / excludeKeywordsCase-insensitive keywords matched in the title and summary
publishedAfterOnly items published after a date or a relative time (24 hours, 7 days, 2 weeks)
includeContentAdd contentHtml (the full HTML content from the feed)
useAllDiscoveredFeedsMonitor every feed a website advertises (posts, podcast, categories), not only the main one
discoverOnlyFeed finder mode: list the feeds, don't read the items

Example: daily new-posts monitor, filtered

{
"startUrls": ["https://wordpress.org/news/", "https://github.com/apify/crawlee/releases.atom", "https://www.youtube.com/@GoogleDevelopers"],
"onlyNewItems": true,
"emitOnBaseline": false,
"includeKeywords": ["release", "security", "AI"]
}

Create a schedule (Console > Schedules), then add an integration (Slack, email, Zapier, Make or a webhook) that fires on each successful run.

Example: find the feeds of many websites

{ "startUrls": ["github.blog", "https://www.theverge.com/", "daringfireball.net"], "discoverOnly": true }

Output

One row per item. This is a real example (YouTube channel feed found from an @handle URL):

{
"feedUrl": "https://www.youtube.com/feeds/videos.xml?channel_id=UC_x5XG1OV2P6uZZ5FSM9Ttw",
"feedTitle": "Google for Developers",
"siteUrl": "https://www.youtube.com/channel/UC_x5XG1OV2P6uZZ5FSM9Ttw",
"feedType": "atom",
"guid": "yt:video:vLAegkGGWrM",
"title": "Building with Gemini in Google Antigravity",
"link": "https://www.youtube.com/watch?v=vLAegkGGWrM",
"published": "2026-11-12T17:00:11.000Z",
"updated": "2026-11-12T17:04:24.000Z",
"author": "Google for Developers",
"summary": "See what you can build using our latest Gemini models in Google Antigravity. From in-browser animation and multimodal prototyping to autonomous codebase modernization, ...",
"categories": [],
"enclosure": null,
"imageUrl": "https://i3.ytimg.com/vi/vLAegkGGWrM/hqdefault.jpg",
"isNew": true,
"firstSeenAt": "2026-11-16T09:38:22.029Z",
"inputUrl": "https://www.youtube.com/@GoogleDevelopers",
"scrapedAt": "2026-11-16T09:38:22.029Z"
}

Podcast items include the audio file:

"enclosure": { "url": "https://wordpress.org/news/podcast-download/17684/episode-86-my-first-wordpress-experience.mp3", "type": "audio/mpeg", "length": 10013967 }

In feed finder mode each row has inputUrl, feedUrl, feedTitle, feedType (rss / atom / rdf / json), discoveredVia (direct / link-tag / common-path / youtube), itemCount, latestItemTitle and latestItemPublished.

A per-feed run summary (items in the feed, new, filtered, HTTP 304, errors) is saved to the key-value store record RUN-SUMMARY. Feeds that fail are listed there and in the log, not in the dataset, so you never pay for an error row. If no feed at all can be read (site down, blocked or rate-limited), the run still finishes normally with a status message saying so, so check the status message or RUN-SUMMARY when a scheduled run returns nothing.

Empty runs leave one free status row. When a run outputs no items (nothing new since the last run, a silent baseline, or no feed readable), the dataset gets exactly one row with "type": "run-status", a status (no-new-items, baseline-recorded, no-feed-readable, no-feeds-found) and a human-readable title. It is never charged. Item rows never have this type, so filter it out in your integration, or set statusRowWhenEmpty to false to get an empty dataset instead.

Pricing

Pay per result: you pay a small fee per run plus a fee per item output (or per feed found in feed finder mode). Already-seen items, filtered items and silent baselines are free. Set a maximum charge per run in the run options to cap your cost.

Limits (honest)

  • A feed shows only its latest N items (often 10 to 50). If more than N posts are published between two runs, the older ones fall out of the feed before the Actor can see them. Schedule more often for busy feeds.
  • "New" means the item's identity (guid, else link, else title + date) was not seen before. A publisher that changes guids re-publishes items as new; edited items with the same guid are not reported again.
  • Discovery reads one page per URL plus at most 6 common paths. Sites with no feed at all return a clear error for that URL (this is not a website-to-RSS generator).
  • YouTube @handle URLs need one plain fetch of the channel page. If YouTube shows a consent or bot page, the Actor says so: use the youtube.com/channel/UC... URL instead.
  • Feeds behind logins, paywalls or bot protection (Cloudflare challenges) are not supported. No proxies are used.
  • Monitoring state lives in the named key-value store rss-feed-monitor-state in your account. Changing stateKey starts a fresh baseline.

Reads public feeds that publishers provide for automated consumption, plus one public page per URL for discovery, with a clear user agent. Author names are output only as the feed publishes them. Respect each publisher's terms when you republish content. Not affiliated with YouTube, Google, GitHub, WordPress or any publisher whose feed it reads.

More from Lindenbyte

See all our Actors at apify.com/lindenbyte.