How to Find the RSS Feed for Any Website (The 2026 Definitive Guide)
A deep, tested guide to finding the RSS feed for any website — with an interactive feed finder, a 50+ platform pattern cheatsheet, troubleshooting for JavaScript-rendered sites, and working methods for sites that seem to have no feed at all.
Every existing guide to finding RSS feeds tells you the same three things: append /feed to the URL, view the page source and search for "rss," or install a browser extension. That advice was written in 2014. It works for the WordPress blogs those articles were written about, and it fails the moment you try to follow a YouTube channel, a Substack, a Mastodon account, a GitHub repo, a subreddit, or a site rendered by JavaScript.
This guide is different. It covers every major platform, every edge case, and every fallback, and it does it with an interactive tool that actually finds feeds for you instead of describing the process in the abstract.
Paste a URL below and try it.
Feed Finder
Paste any website. We scan 20+ locations in parallel.
Under the hood, this tool runs five discovery methods in parallel and validates every candidate against a live RSS parser. It's the same logic we use internally at Nutshell to detect feeds when users add a new source. If it finds a feed here, your reader can find it too.
The rest of this article explains why those methods work, which method to reach for first for each type of site, and what to do when a site appears to have no feed. If you're in a hurry, skip to the platform cheatsheet — it covers 50+ platforms with copy-paste URL patterns.
Why this guide exists
We built this because we kept hitting the same ceiling in other tutorials: they stop at WordPress. In 2026, "every website" means blogs, newsletters, video platforms, federated social networks, dev tools, academic databases, and single-page apps. Each needs a different approach, and using the wrong one wastes hours.
Start here: the 30-second decision tree
Pick the type of site you're trying to follow, and we'll jump you to the fastest method for that specific platform. This alone will solve 90% of feed-finding problems without you reading anything else.
How websites actually advertise their feeds
Before we dive into methods, understand what's really happening. When a site publishes an RSS feed, it signals this to readers, browsers, and tools through a single HTML tag in the page's <head>: the autodiscovery link.
<link rel="alternate" type="application/rss+xml" title="Ars Technica - All content" href="https://feeds.arstechnica.com/arstechnica/index/" />
relMarks this as an alternate representation of the page.typeThe MIME type — can also be application/atom+xml.titleHuman-readable label shown in feed-discovery UIs.hrefThe actual feed URL — this is what you copy.This is the canonical way to advertise a feed, and it's what every RSS reader, browser extension, and feed-finder tool actually looks for. It's been the W3C-recommended pattern since 2002. When you "paste a URL into your RSS reader and it just works," this tag is why.
Every method in this guide either reads this tag directly, guesses at common URL patterns when the tag is missing, or works around the absence of a feed entirely. Once you internalize that, the methods stop feeling like tricks and start feeling like a decision hierarchy:
- Read the
<link rel="alternate">tag (the right answer). - If it's missing, guess common URL patterns (almost always works for blogs).
- If that fails, use a platform-specific URL recipe (for YouTube, Reddit, etc.).
- If the platform has no native feed, use a third-party bridge (RSSHub, OpenRSS).
- If none of that works, generate your own feed from the HTML (PolitePol, Feedbin newsletters, or Nutshell).
That's the entire landscape in five steps. Every other "tip" you've read is a variant of one of these.
Method 1: Paste the homepage into your RSS reader
This is the method nobody mentions because it's too obvious, but it's the one you should try first.
Modern RSS readers — Feedly, Inoreader, Feedbin, NetNewsWire, Reeder, and Nutshell — all implement autodiscovery. You paste the homepage URL (https://arstechnica.com, not https://feeds.arstechnica.com/...) and the reader fetches the page, parses the <link rel="alternate"> tag, and subscribes you to the feed it finds.
Why this beats every other method
Your RSS reader already has network access, an HTML parser, and a feed validator. Making it do the work yourself is strictly slower and introduces failure modes (typing the URL wrong, copying the excerpt feed instead of full-text, etc.). Try the homepage first. Always.
This works for roughly 70% of blogs, news outlets, and personal sites in 2026. It fails in three specific cases, which we'll handle in the next sections:
- Sites with multiple feeds (categories, authors, tags) — you need to pick which one.
- Sites that don't advertise autodiscovery — you need to guess the URL pattern.
- Platforms with no HTML homepage to scan — YouTube, Reddit, Mastodon, etc.
Method 2: Try the URL pattern
For 43% of the web — every WordPress site — the feed is at /feed. For Substack, it's also /feed. For Ghost, it's /rss. For Blogger, it's /feeds/posts/default. Each blogging platform has a convention, and once you know them, you can find the feed without even loading the homepage.
This is the single most useful skill you can learn for RSS. Memorize the top four patterns and you'll find 80% of feeds in under five seconds:
WordPress, Substack, Beehiiv, Hashnode — the biggest four. Try this first.
When /feed doesn't work, try /rss, then /feed.xml, then /atom.xml. When none of those work, you're almost certainly on a non-blog platform and you need a specific pattern.
The platform pattern cheatsheet (50+ platforms)
Search for the platform you're trying to follow. Every pattern below is tested and works as of April 2026.
WordPress
Blogs & newsletters{site}/feedAlso: /feed/atom, /category/NAME/feed
Medium (publication)
Blogs & newslettershttps://medium.com/feed/{publication}Medium (user)
Blogs & newslettershttps://medium.com/feed/@{username}Dev.to (tag)
Blogs & newslettershttps://dev.to/feed/tag/{tag}Blogger
Blogs & newsletters{site}/feeds/posts/defaultSquarespace
Blogs & newsletters{site}/{page}?format=rssYouTube (channel)
Videoyoutube.com/feeds/videos.xml?channel_id={UCxxx}Need channel ID, not @handle
YouTube (playlist)
Videoyoutube.com/feeds/videos.xml?playlist_id={PL…}Twitch
Videotwitchrss.appspot.com/vod/{username}Unofficial third-party
Vimeo (channel)
Videovimeo.com/channels/{channel}/videos/rssPeerTube
Video{instance}/feeds/videos.xml?accountId={id}Threads
Socialopenrss.org/threads.net/@{user}Via OpenRSS; Threads has no native feed
X / Twitter
Social(use rss.app or openrss.org)No native feed since 2013
(use rssbridge or openrss.org)Breaks often; IG blocks scrapers
GitHub (repo releases)
Developergithub.com/{owner}/{repo}/releases.atomGitHub (repo commits)
Developergithub.com/{owner}/{repo}/commits/{branch}.atomGitHub (repo tags)
Developergithub.com/{owner}/{repo}/tags.atomStack Overflow (tag)
Developerstackoverflow.com/feeds/tag?tagnames={tag}Stack Overflow (user)
Developerstackoverflow.com/feeds/user/{id}PyPI project
Developerpypi.org/rss/project/{name}/releases.xmlReddit (search)
Communityreddit.com/search.rss?q={query}Lemmy
Community{instance}/feeds/community/{name}.xmlApple Podcasts
Podcasts(use castos.com/tools/find-podcast-rss-feed)Apple hides the real feed
Spotify
Podcasts(use castos.com/tools/find-podcast-rss-feed)Spotify hides the real feed
Overcast
Podcasts(use the 'Share' sheet for RSS)Google Scholar (alert)
Academic(create alert → RSS link in email)Use Scholar Alerts Chrome extension
PubMed (search)
Academicpubmed.ncbi.nlm.nih.gov/rss/search/{id}/?limit=...Semantic Scholar
Academic(no native RSS; use RSSHub /semanticscholar)Google News (search)
Searchnews.google.com/rss/search?q={query}Google News (topic)
Searchnews.google.com/rss/topics/{id}The New York Times
Newsrss.nytimes.com/services/xml/rss/nyt/{section}.xmlBBC News
Newsfeeds.bbci.co.uk/news/{section}/rss.xmlReuters
News(use openrss.org/reuters.com/{section})Dropped native feeds in 2020
Ars Technica
Newsfeeds.arstechnica.com/arstechnica/{section}Bookmark this
If you read a lot on any of these platforms, bookmark this page and come back when you need a pattern. It's the most complete public list we've found, and we update it quarterly.
Method 3: View the page source
When autodiscovery fails and pattern-guessing fails, the site probably has a feed hidden somewhere. View-source is how you hunt for it.
On any page, right-click and pick View Page Source (or press Cmd+U on Mac, Ctrl+U on Windows). Search the raw HTML (with Cmd+F / Ctrl+F) for any of these keywords, in roughly this order:
rss— catches most WordPress, Ghost, and news sitesatom— catches GitHub, Mastodon, and anything using the modern Atom standardfeed— catches generic patterns and custom feedsapplication/rss+xmlorapplication/atom+xml— the MIME types
If you find a hit inside a <link rel="alternate"> tag, copy the href value. That's your feed URL.
JavaScript-rendered sites break this method
If the site is a single-page app (SPA) built with React, Vue, or similar, the page source won't contain the rendered HTML — it'll be mostly empty with a single <div id="root">. The autodiscovery tag, if it exists, is injected by JavaScript after page load. In this case, use your browser's DevTools (F12) → Elements tab instead of View Source. The Elements tab shows the live, post-render DOM.
Even with DevTools, plenty of modern sites simply don't expose a feed, because the developer never added one. When that happens, skip to Method 5 or the workarounds section.
Method 4: Use a browser extension
Browser extensions are slower than the Feed Finder above and slower than pattern-guessing, but they win in one specific scenario: browsing-to-feed conversion. You're reading an article, realize you want to follow the site, and want to grab the feed without switching tabs.
Three extensions that actually work in 2026:
- Feedbro (Chrome, Firefox): Auto-detects feeds on any page, shows them in a popup. Also works as a full-featured reader.
- RSS Subscription Extension (Chrome): Reads autodiscovery tags and shows matching readers. Minimal, no account needed.
- Want my RSS (Firefox): The Firefox-native equivalent, restored after Mozilla removed built-in RSS support in 2018.
If you switch browsers often, skip the extension route and use the Feed Finder above — it works the same on every device without install.
Method 5: Third-party feed bridges
Some platforms — X, Instagram, Facebook, Threads, and any site whose owner just never added RSS — need an intermediary. A feed bridge is a service that scrapes a platform and generates an RSS feed you can subscribe to.
Four bridges worth knowing in 2026:
| Bridge | Covers | Hosting | Reliability | Best for |
|---|---|---|---|---|
| RSSHub | 300+ sites (X, IG, TikTok, YT, academic, Chinese social) | Self-host or public | High when self-hosted | Power users who want control |
| RSS-Bridge | ~200 sites, heavy on European outlets | Self-host | Medium (scrapers break) | Privacy-conscious self-hosters |
| OpenRSS | Common platforms, hosted for free | SaaS (free) | Medium-high | Quick one-offs, no setup |
| rss.app / fetchrss | Any URL (creates custom feed) | SaaS (freemium) | High | Sites with no feed at all |
For X/Twitter, OpenRSS (openrss.org/twitter.com/{username}) is the fastest. For Instagram, the official API restrictions mean most bridges are flaky — RSSHub self-hosted is the only consistent option. For mainstream news sites that dropped RSS (Reuters, some local newspapers), OpenRSS usually has them.
Which method works for which site
Now that you know all five methods, here's the cheat sheet for which to reach for first. Hover any column for a refresher on how that method actually works.
Which method works for which site
Hover a column for details. Based on observed behavior across 9 common site types.
| Site type | Autodiscovery | URL pattern | View source | Browser extension | Third-party bridge |
|---|---|---|---|---|---|
| WordPress blog | Reliable | Reliable | Usually | Usually | Flaky |
| Substack | Usually | Reliable | Mixed | Usually | Mixed |
| News outlet | Usually | Mixed | Mixed | Usually | Mixed |
| Ghost / Hashnode | Reliable | Reliable | Usually | Usually | Flaky |
| YouTube channel | No | Reliable | Mixed | Flaky | Mixed |
| Reddit / Mastodon | No | Reliable | Rare | Flaky | Usually |
| Corporate site (SPA) | Rare | Rare | Rare | Rare | Usually |
| X / Instagram | No | No | No | No | Reliable |
| Academic (arXiv / PubMed) | Flaky | Reliable | Flaky | Flaky | Mixed |
The takeaway: autodiscovery and URL patterns are your workhorses for anything content-y — blogs, news, newsletters, Ghost/Hashnode sites, academic sources. Third-party bridges are the only viable option for social platforms and SPAs. Browser extensions are a quality-of-life tool, not a discovery strategy.
Platform deep-dives: the tricky ones
Most platforms are easy. A handful trip people up consistently. Here's what you actually need to know for each.
YouTube: you need the channel ID, not the handle
YouTube feeds require the UC... channel ID in the URL. The problem: YouTube's modern URLs use @handle format, which is not the channel ID.
To convert a handle to a channel ID:
- Visit the channel page (
youtube.com/@handle). - View page source, search for
"channelId":"UC. - Copy the 24-character string starting with
UC. - Build the feed URL:
https://www.youtube.com/feeds/videos.xml?channel_id=UCxxxxxxxxxxxxxxxxxxxxxx.
Third-party converters like commentpicker.com/youtube-channel-id.php do the same thing in one click. The Feed Finder above handles this automatically — just paste the channel URL.
Reddit: append .rss and expect rate limits
Every Reddit URL supports .rss. Subreddits (/r/selfhosted/.rss), users (/user/spez/.rss), search results (/search.rss?q=...), and even specific threads all work.
The catch: unauthenticated RSS requests are aggressively rate-limited. If your reader starts getting 429 errors, you need to authenticate. Most self-hosted readers (FreshRSS, Miniflux) support Reddit OAuth. SaaS readers like Feedly and Inoreader handle this transparently.
Substack: works, but check the full-text status
Every Substack newsletter exposes /feed, but paid newsletters deliver excerpts only to the RSS feed. The full content is paywalled. If you're a paid subscriber and want full-text RSS, some newsletters offer a private feed URL in the subscriber settings — check there before assuming it's impossible.
Mastodon & Bluesky: the fediverse is RSS-native
Mastodon profiles (mastodon.social/@user.rss) and Bluesky profiles (bsky.app/profile/handle/rss) expose feeds by default with no auth required. This is part of the fediverse's open-by-default design, and it's why so many former-Twitter users are rediscovering RSS in 2026.
GitHub: four feeds per repo
GitHub exposes Atom feeds for almost everything. For any repo, you can follow:
- Releases:
github.com/{owner}/{repo}/releases.atom— what most people actually want - Tags:
github.com/{owner}/{repo}/tags.atom— all Git tags, including pre-releases - Commits:
github.com/{owner}/{repo}/commits/{branch}.atom— high-volume, useful for forks - User activity:
github.com/{username}.atom— PRs, issues, stars
For tracking dependencies, releases.atom is the gold standard. Subscribe to the ten most important packages in your stack and you'll know about security releases before your package manager does.
Podcasts: Apple and Spotify hide the real feed
Every podcast is RSS. Apple Podcasts and Spotify are just directories layered on top of RSS feeds. But both hide the underlying feed URL.
To get it:
- Paste the Apple Podcasts or Spotify URL into
castos.com/tools/find-podcast-rss-feed. - Or find the podcast's website and look for the RSS icon in their footer (most have one).
- Or view-source on the podcast's episode player and search for
enclosure— the feed URL is usually nearby.
The resulting feed works in any RSS reader and any podcast app.
Academic: arXiv, PubMed, and Google Scholar
For keeping up with research, three patterns:
- arXiv: every category has a feed at
export.arxiv.org/rss/{category}. Example:export.arxiv.org/rss/cs.AIfor AI papers. - PubMed: run a search on
pubmed.ncbi.nlm.nih.gov, then click the "Create RSS" button above the results. PubMed generates a permanent search URL. - Google Scholar: no native RSS. Workaround: create a Scholar Alert, then use the Scholar Alerts Chrome extension to convert the email alert into an RSS feed.
What if the site doesn't publish a feed at all?
Roughly 25% of modern commercial sites — marketing pages, corporate blogs built as SPAs, news sites that stripped out RSS during a redesign — simply don't offer a feed. You have four options, in order of preference.
Option 1: PolitePol or rss.app
PolitePol (politepol.com) and rss.app are "web-to-RSS" services. You give them a URL, point them at the repeating elements on the page (the list of articles, blog posts, whatever), and they generate a feed that updates automatically. PolitePol is free and open-source. rss.app is commercial but has the better UX.
This is the right choice for corporate sites, marketing blogs, and any page that looks like a feed but doesn't publish one.
Option 2: RSSHub or RSS-Bridge
If the site is on a common platform (X, Instagram, TikTok, YouTube Shorts, Chinese social media), chances are RSSHub already has a route for it. Check docs.rsshub.app/en/ and search for the platform. If you find it, you can either use the public instance (rsshub.app) or self-host for reliability.
RSS-Bridge is similar, with heavier European coverage. Both are open-source.
Option 3: Feedbin Newsletters (for email-only content)
For writers who only publish to email (many substacks, corporate newsletters, personal mailing lists), you can subscribe using a dedicated address provided by services like Feedbin Newsletters or Kill the Newsletter. They turn incoming emails into RSS items, which flow into your reader like any other feed.
Option 4: Let Nutshell handle the whole stack
If you're trying to follow 20+ sources and the feed-hunting is exhausting, we built Nutshell for exactly this moment. You add whatever sources you can — URLs, feeds, YouTube channels, newsletters — and we handle discovery, deduplication, and daily AI-summarized digests delivered to your inbox. You stop managing a reader and start reading.
Stop hunting for feeds. Let us do it.
Nutshell detects feeds from any URL, summarizes them daily with AI, and delivers a single morning digest. Free to try.
Try Nutshell freeValidate that the feed actually works
Finding the URL is half the battle. The other half is making sure it's alive and actually contains what you think it does.
Three quick checks:
- Paste the URL into your browser. If you see an XML document with
<rss>or<feed>at the top, you have a valid feed. If you see a 404, the feed doesn't exist. If you see a login wall, the feed is auth-gated. - Use the W3C Feed Validator (
validator.w3.org/feed). It catches malformed feeds, missing timestamps, and encoding issues — the kinds of problems that cause feeds to silently fail in some readers and work in others. - Check the publish date of the last item. A feed whose most recent item is from 2022 is probably abandoned. Every reputable reader shows this as an unread indicator, but for your first pass, just eyeball the first item's date.
The Feed Finder at the top of this article does all three checks automatically and only returns feeds that parse cleanly.
Import the feeds you found
Once you have the feed URL, adding it to your reader takes seconds. Two patterns:
Single feed: In any RSS reader, hit "Add feed" or "Subscribe" and paste the URL. Every reader supports this.
Bulk import (OPML): If you have a list of feeds, save them as an OPML file (the standard export format for RSS readers) and import the whole bundle at once. Most readers expose OPML import under Settings → Import/Export. This is how you migrate from one reader to another — Feedly → Inoreader, for example — without losing a year of curation.
Quick OPML template
OPML is just XML with an <outline> tag per feed. You can hand-build a file with nothing more than a text editor and a list of feed URLs. Every reader reads the same format.
Troubleshooting: common failure modes
When the Feed Finder comes up empty, the issue is almost always one of these four:
1. The site is a pure SPA. The homepage is empty HTML; content is loaded by JavaScript. Autodiscovery tags are injected post-render. Use DevTools → Elements instead of View Source, or check if the site is on a common platform and use a pattern.
2. The feed exists but is on a subdomain. Many sites publish feeds at feeds.example.com/... rather than example.com/feed. Our tool catches most of these through autodiscovery, but if the site doesn't advertise, you may need to guess subdomains.
3. The feed is behind Cloudflare or a WAF. Corporate sites sometimes block the user-agents RSS readers send. Some readers let you customize the user-agent; FreshRSS, Inoreader, and Feedly all expose this setting.
4. The feed requires authentication. Private podcast feeds (Patreon, premium subscriptions) include a token in the URL. If you're a subscriber, the feed URL is usually in your account settings or your welcome email.
If you hit none of these and still can't find a feed, the site genuinely doesn't publish one. That's when you fall back to PolitePol, RSSHub, or Nutshell.
FAQ
Does every website have an RSS feed?
No. It's a common myth. Every WordPress site does (43% of the web). Every Substack, Ghost, and Hashnode site does. Every YouTube channel, subreddit, Mastodon profile, and GitHub repo does. But single-page apps, most modern SaaS marketing pages, and walled-garden social platforms (X, Instagram, TikTok) don't publish native feeds. Total coverage is probably around 70–75% of the content web.
Can I find the RSS feed for a specific author or topic on a site?
Usually. WordPress exposes /author/{name}/feed and /tag/{name}/feed. Reddit exposes search feeds (/search.rss?q=...). Google News exposes search feeds (/rss/search?q=...). Most academic databases do too. For sites that don't, use the "filter by author/tag" feature in your RSS reader on the full feed.
Why does the feed I found only show excerpts?
Because the publisher chose to. Many paywalled sites (Ars Technica, The Atlantic) publish excerpt-only feeds on their free tier and full-text feeds for paid subscribers. This is a deliberate choice, not a bug. If you want full-text, subscribe or use a service like Feedbin that fetches full-text server-side (within the terms of service the publisher sets).
Is RSS being phased out?
Exactly the opposite. Adoption has climbed every year since 2020, the fediverse is RSS-native, and every major LLM-powered reader in 2026 is built on RSS under the hood. The tech press keeps declaring RSS dead; the installed base keeps growing. (For a longer version of this argument, see our piece on why RSS still matters.)
What's the difference between RSS and Atom?
Functionally: nothing you need to care about. Both are XML feed formats. Atom was designed in the mid-2000s as a more rigorously specified successor to RSS 2.0. Every modern RSS reader parses both transparently. When you see .atom in a URL (GitHub feeds, for example) it's Atom; when you see .rss it's RSS. Same thing to you.
Can I make an RSS feed for a site I don't own?
Yes, using a web-to-RSS service like PolitePol, rss.app, or RSSHub. You're scraping public content, which is legal in most jurisdictions for personal use. Be a good citizen about rate limits and don't republish the content elsewhere.
The short version
If you read nothing else from this article, remember this:
- Paste the homepage into your RSS reader first — it almost always just works.
- When it doesn't, try
/feedthen/rss— covers 80% of blogs. - For platforms (YouTube, Reddit, GitHub, Mastodon), use the specific URL pattern from the cheatsheet above.
- For sites with no feed, use PolitePol, RSSHub, or just let Nutshell handle everything.
Or — easier — scroll back to the top and use the Feed Finder tool. It does the first three automatically, and if it comes up empty, you know it's time for a workaround.
Turn any set of feeds into a daily digest
Nutshell auto-detects feeds from any URL, summarizes them with AI, and delivers a single morning email. No reader to manage, no inbox to clear.
Start free