Broken Link Checker

Broken Link Checker requests every link a page points at and reports which destinations fail.

A dead link costs twice over. The visitor hits a wall and usually leaves, and the crawler spends part of its budget for your site on a URL that returns nothing. Neither problem is visible in the markup — the href looks perfectly correct — so the only way to find them is to request each destination and read what comes back. That is what this does.

Requests use HEAD where the server allows it, since the body is irrelevant. Plenty of hosts answer HEAD with 403 or 405 while serving the same URL happily over GET, so those are retried with GET rather than reported as broken. 401, 403 and 429 responses are separated out too: they are usually bot protection or a login wall, not a dead link.

Working links that redirect are flagged separately. They are not broken, but each redirect is an extra round trip on every click, and pointing the href at the final URL is free.

Two honest limits. Up to 50 distinct destinations are checked within a 25-second budget, so the tool cannot be used to hammer a third-party site — if a page has more links you are told the check was capped. And only links in the server-sent HTML are seen, so anything a script injects after load is invisible here, exactly as it is to a crawler on first pass.