Skip to content

Broken Link Checker

This broken link checker fetches a page you give it, finds every link on it, internal and external, and checks each one live so you can see which return an error, which redirect, and which are healthy, with the status code for each.

Starts from this page, finds up to 19 more on the same site through its sitemap, and checks every distinct link once. Slower than one page: allow up to half a minute.

Note

We fetch the page server-side, extract every link on it (internal and external, up to 40), and check each one live with redirects left unfollowed so its own status code is reported rather than wherever it eventually lands. The up-to-20-pages scope reads more pages from the same site through its XML sitemap, honours robots.txt, checks each distinct link once (up to 120) and lists every page it was found on. A 429 response is flagged separately from a broken link: it means the server wants us to slow down, not that the page is dead.

Switch to the up-to-20-pages scope and it reads more pages from the same site, found through its sitemap, and reports every page each broken link was found on.

Use this tool often? Bookmark this page, so it is one click away next time.

Email me this result

Want a copy for your records or a client? Leave your email and we'll send you this result. It is written by a real person at Crawl Cove, not an automated blast, so it may take a little while to arrive. We'll only use your address to send it, unless you also tick the box below. Privacy policy.

That email address doesn't look right. Please check it and try again.

Something went wrong saving your address. Your result above is unaffected; please try again.

Got it. We'll email your result to you. Nothing else lands in your inbox unless you ticked the tips box.

What this broken link checker shows you

Paste a URL and the page is fetched server-side. Every link on it is extracted, internal and external alike, deduplicated, and checked live, up to 40 per page. Each one comes back as OK (a 2xx response), a redirect (3xx, reported with its target rather than followed further), broken (4xx or 5xx), or unreachable, with the status code shown for every link that returned one. Broken results are listed first.

Checking up to 20 pages at once, with the page each broken link sits on

The one-page scope answers the question you already have about one URL. The up-to-20-pages scope answers the one you usually have next: where else on the site is this link? It starts from the page you enter, reads the site's XML sitemap to pick up to 19 more pages on the same host (falling back to the internal links on your page when there is no sitemap), honours robots.txt, and pools every distinct link found across those pages, checking up to 120 of them once each. Every result then lists the pages it was found on, with the link text used on each, so one broken URL linked from a footer shows up as one row with twenty sources rather than twenty separate problems. The CSV carries the same source list. Links that point at a page already read are not requested twice; their status comes from that read. The run has a time budget, and if a slow host uses it up the remaining links are marked unchecked rather than guessed at.

Why a 429 is reported separately, not as broken

HTTP 429 means the server is asking whoever is asking to slow down. It says nothing about whether the page itself works, and treating it as a dead link is a real mistake we have made and fixed in our own product: measured on our own site (5 Sept 2026), a crawl that ran into the host's own rate limit had six healthy pages misreported as broken, an exact match between the rate-limited requests and the false positives. This checker flags 429 as its own outcome rather than folding it into "broken", and suggests trying again in a minute.

Redirects are not broken links, but they are not free either

A link returning a 3xx is not dead, the page it points to still exists, but every redirect it survives costs a moment of latency and, on an internal link, is usually one edit away from pointing straight at the real destination. This tool reports the redirect and its target without chasing the rest of the chain; if you already know a specific URL redirects and want every hop with its own status code, our redirect checker does that.

Redirect Checker

Up to 20 pages here; your whole site there

Checking the page you already suspect is useful, but the links that quietly rot are usually the ones nobody rechecks: an outbound citation to a page that moved, an old internal link nobody noticed after a restructure. This tool checks both directions on one page, or on a sample of up to 20. Crawl Cove's whole-site broken link check goes further on the internal side, crawling every page and matching every internal link against what it actually found, with no cap once Max pages is raised, and when the crawl ends it checks up to 300 outbound links, the most linked first, for pages that are gone (404 or 410) or domains that no longer exist.

Broken link checker (whole-site) · Full site audit

Frequently asked questions

What counts as a broken link?
A link that returns a 4xx or 5xx status code (404 Not Found and 410 Gone are the most common), or one that could not be reached at all. A redirect (3xx) is reported separately, since the destination still exists.
Is this broken link checker free?
Yes. It is completely free and needs no sign-up. Enter a URL and every link on that page is checked server-side.
Does this check external links too?
Yes, both internal and external links on the page you enter are checked. Crawl Cove's own broken link check covers the internal links on every page it crawls, up to Max pages, and when the crawl ends it checks up to 300 external links for pages that are gone (404 or 410) or domains that no longer exist.
How many links does it check?
In the one-page scope, up to 40 links on that page, in the order they appear. In the up-to-20-pages scope, up to 120 distinct links across all the pages read, broken ones listed first. Past either cap the result says how many were found and how many were checked.
Does this check my whole site?
Not all of it. The one-page scope checks the links on the page you enter. The up-to-20-pages scope reads that page plus up to 19 more from the same site, taken from its XML sitemap in the order listed there, or from the internal links on your page when no sitemap is found. Crawl Cove crawls a whole site and matches every internal link against the pages it actually found there, with no link-count cap once Max pages is raised. Max pages ships at 500 a crawl and you can raise it to 100,000.
How does the up-to-20-pages scope choose which pages to read?
It looks for the site's XML sitemap the way a crawler would: the Sitemap lines in robots.txt first, then /sitemap.xml, /sitemap_index.xml and /wp-sitemap.xml. The first 19 same-host URLs listed there, after the page you entered, are read. If no sitemap turns up, the internal links on your page are used instead. Pages that robots.txt disallows for all crawlers are listed as skipped, not read. To steer which pages are included, enter a page deeper in the site or point the tool at the section you care about.
What does "unchecked" mean on a result?
The run has a time budget so that a slow site cannot stall the page. When the budget runs out before every pooled link has been requested, the remaining links are marked unchecked rather than guessed at, and the summary says the result is partial. Running it again a minute later usually finishes, because pages and links that answered quickly the first time are not the ones that were slow.
Why does this run on your server instead of my browser?
A browser cannot reliably fetch pages on other domains itself (that is what CORS exists to prevent), and even where it can, it cannot read a clean status code the way a direct server-side request can. Checking server-side is the only way to get a real status for every link in one pass.
What does a 429 result mean?
The server is asking us to slow down, which is not the same as the link being broken. The page behind it is very often fine a moment later. This checker reports 429 as its own outcome rather than counting it as broken.
What is the difference between this and the Redirect Checker?
This checks whether every link on a page works at all, in one pass. The Redirect Checker follows one URL you already suspect redirects, through every hop of its chain, with each hop's own status code. Use this one to scan a page; use that one once you have found a specific redirecting link and want the full path.
Can I check a page I do not own?
Yes, the same as any browser could reach the URL directly. It is a read-only check: one request per link, nothing is submitted to any target, and results are not stored anywhere you or anyone else can browse back to.
How do I find a broken link on the page once I know its URL?
Each result shows the link text the page uses for it (or the image alt text, aria-label or icon title when the link has no words) and how many times the page links that URL. For every link that is not OK it also shows the link's opening tag exactly as it appears in the page's HTML, and, for one page, the line it starts on: open View source (Ctrl+U), then search for that tag or jump to the line. The line refers to the HTML the server sends, so on a page whose links are added by JavaScript after it loads, search the rendered page for the link text instead. Search for the URL in your CMS to find every place to fix. The CSV export includes all of it.
Can I export the results?
Yes. A "Download CSV" button appears once a check completes, listing every link's URL, link text, how many times it was linked, verdict, status code, whether it is internal or external, any note and, in the one-page scope, the line and opening tag in the page source, or in the up-to-20-pages scope, every page it was found on with the line on each.

From one page to your whole site

This checks one page, or a sample of up to 20 pages from one site, including the links pointing off them. Crawl Cove's own broken link check goes further: it crawls your whole site and matches every internal link against the pages it actually found there, with no link-count cap once Max pages is raised, and when the crawl ends it checks up to 300 outbound links for pages that are gone or domains that no longer exist. Use this free tool for a quick check of one page. See check broken links across your whole site or compare the plans.

Download free