RSS and Atom Feed Reader

What it does

This Actor reads news and blog feeds and returns every item as a clean row: title, link, publish and update dates in UTC, author, categories, a plain-text summary and any attached files (such as podcast audio). Give it one feed or a list of them.

It reads RSS (0.9x, 1.0 and 2.0), Atom and JSON Feed. If you give it a website instead of a feed, for example a blog's home page, it finds the feed the page announces and reads that.

You can keep only items published after a date, and you can ask for the full text and HTML of each item when the feed carries it.

It reads each site's robots.txt before anything else and obeys it. It never logs in, never sends a cookie and never tries to get past a CAPTCHA. Attached files are listed, not downloaded. Email addresses are removed from item titles, summaries, content, authors, categories, ids, links and attachments before anything is saved; text that only looks like an address, such as react@18 or logo@2x.png, is kept.

How to use it

1. Open the Actor in Apify Console.
2. Put one feed or website address per line in Feed or website URLs.
3. Set Maximum items per feed, and Published on or after if you only want recent items.
4. Click Start.
5. When the run ends, open the Output tab. The Items view shows each item with its feed, date, title, link and summary. Export as CSV, JSON or Excel.

Input

Field	What it does	Default	Limits
Feed or website URLs (feedUrls)	The feeds to read, one address per line. A web page works too, if it announces a feed.	None. Prefilled with one public feed	http and https only. Up to 500 lines. Duplicates are removed
Maximum items per feed (maxItemsPerFeed)	The most items saved from each feed, in the order the feed lists them (usually newest first).	50	1 to 1,000
Published on or after (publishedAfter)	Only items dated on or after this day, in UTC: 2026, 2026-06, 2026-06-30 or a full time. Items with no date are left out when this is set.	None
Include full content (includeContent)	Also return the full HTML and text of each item, when the feed carries it.	false	Each is cut at 200,000 characters
Download timeout (requestTimeoutSecs)	How long a site may take to start answering. Reading the answer may then take up to four times this, so one download can take up to five times this, for each redirect.	30 seconds	5 to 120 seconds

A date that cannot be read (including a day that does not exist, or a time that is not in UTC) stops the run before anything is requested, so nothing is charged for items you did not ask for. 2026-1-1 is read as 1 January.

The same feed twice. A feed reached a second time, through another address, a redirect or a page that announces it, is read once; the other lines are listed in SKIPPED as duplicate_feed. Two feeds count as the same when they have the same address, or the same format, title and item titles and dates. Different feeds that share items (a site's main feed and one of its category feeds, say) are separate, so a shared item can appear once for each feed, and each is charged.

Output

One dataset item per feed item:

Field	What it holds
sourceUrl, feedUrl, discoveredFrom	The address you gave, the feed that was read (after redirects), and the page the feed was found on when you gave a page
feedTitle, feedLink, feedFormat	The feed's title, its website, and its format: rss20, rss10, rss090 to rss094, atom10, atom01 to atom03, json1 or json1.1
position	Where the item sits in the feed, from 1
itemId, title, link	The item's id (RSS guid, Atom id or JSON Feed id; an RSS guid that is a relative address is made absolute, like a link; an id that holds an email address is replaced by a stable sha256: code), its title as plain text, and its web address
published, updated	Dates in ISO 8601, UTC. Empty when the feed does not say. In RSS feeds that date items with dc:date, that date is the published date
author	The author name the feed gives, without email addresses
categories	The item's categories or tags
summary, summaryTruncated	The summary as plain text, at most 2,000 characters, and whether it was cut
contentHtml, contentText, contentTruncated	The full content, when Include full content is on and the feed carries it. In the HTML, scripts, styles, frames, embedded objects, event handlers, javascript: and data: links and mailto: links are removed by feedparser's sanitizer and the Actor; images and other markup are kept as the feed has them. Relative links are made absolute. If an email address could still show in the HTML (for example one split across tags), contentHtml is left empty and only contentText is given
enclosures	Attached files: url, type and lengthBytes as the feed declares them
readAt	When the feed was read

Example row from a test run of the prefilled feed (some fields left out):

{"feedTitle": "Apify Blog", "feedFormat": "rss20", "position": 1, "title": "Get more people to use your Actor with task examples", "link": "https://blog.apify.com/actor-task-examples/", "published": "2026-10-22T08:00:00Z", "author": "Fabian Maume", "categories": ["Developer community", "Use cases"], "summaryTruncated": false, "enclosures": []}

Pricing

This Actor is priced per event, with two charges:

• Actor start: Apify's automatic start event, $0.0001 per start (one event per GB of run memory, minimum one; this Actor runs at 256 MB, so one per run).
• Item: one charge for each feed item written to the dataset.

Price per item: $1.00 per 1,000 items, which is $0.001 each.

These are never charged: addresses that could not be read (robots.txt blocks, timeouts, errors, 401, 403, 429, login and CAPTCHA pages), pages that announce no feed, answers that are not a feed, feeds that cannot be read, feeds with no items, items outside the date filter, repeated items within a feed, the same feed reached a second time (see above), and the REPORT and SKIPPED records.

If you set a maximum cost for the run and it is reached, the Actor saves no more items, lists the rest in SKIPPED as spending_limit_reached, and ends the run without an error.

Records

Record	What it holds
REPORT	Why the run stopped, the settings used, feeds read and their formats, items saved per feed, items left out by date, repeated items dropped, robots.txt results and requests per site
SKIPPED	Every line that gave no items, could not be read, or was cut short, with the reason: robots_disallowed, http_error, no_feed_found, not_a_feed, unreadable_feed, no_items, duplicate_feed, too_large, timeout, spending_limit_reached and others. Only the items in the dataset are charged

Rules it follows

• robots.txt is read first on every site, including every site a redirect or a feed link leads to, and obeyed.
• One request at a time to any one site, with the site's Crawl-delay. A Crawl-delay over 30 seconds skips the site.
• Answers of 401, 403 and 429 are listed and not retried. Redirects to login or CAPTCHA pages are not followed.
• On Apify, private and local network addresses are refused.
• Feeds larger than 10 MB are skipped, compressed or not.
• A feed may not declare XML entities: the answer is decoded first (UTF-8, UTF-16 or UTF-32), every DOCTYPE and stray text before its first element is removed, an answer that still declares entities there is refused as not a feed, and no outside DTD or entity is ever fetched.
• One feed that cannot be read is listed as unreadable_feed and never stops the run.
• There is no cookie, header, login, proxy or user agent input. The user agent starts with rss-atom-feed-reader/0.1.

Your responsibility

Feeds are published to be read, but their items belong to their publishers. You are responsible for following each site's terms of use and copyright in how you use and republish what this Actor returns.

Limits

Limit	Value
Feeds per run	1 to 500
Items per feed	1 to 1,000
Feed size	Up to 10 MB
Summary	2,000 characters
Full content	200,000 characters each for HTML and text
Memory	256 MB