Internal extraction
Resolve relative href values against the live page while excluding fragments, email, phone, and external destinations.
Free broken link checker / one public page
Find dead internal links without calling every timeout broken.
Enter one public page. RankLatch extracts its server-rendered internal links and checks a bounded live sample, separating confirmed failures from redirects, restrictions, and requests that could not be verified.
✓ No signup✓ Same-host links✓ Read-only
12maximum live destinations
2redirect hops followed
1source page inspected
$0no account required
Know what failed, what moved, and what simply could not be confirmed.
A timeout, bot restriction, redirect, and 404 are different outcomes with different fixes. RankLatch preserves that distinction so a quick check does not create unnecessary work.
Confirmed failures
Reserve the broken label for 404, 410, and final server-error responses.
Redirect evidence
Show the initial and final status plus destination when an internal link relies on a server redirect.
Honest uncertainty
Keep access restrictions and incomplete requests separate from links proven dead.
A bounded page check—not a whole-site crawler.
- ✓Internal href values present in the source page’s server HTML
- ✓Up to 12 unique same-host destinations in source order
- ✓Initial and final HTTP response plus bounded redirects
- ✓Confirmed failures separated from restrictions and incomplete checks
- —Every link across every page on the website
- —External destinations and client-rendered links
- —Destination relevance, usability, indexability, or ranking value
- —Forms, buttons without href values, downloads, and browser interactions
Current check → page protection
Broken links are often a symptom of a page change.
RankLatch monitors the page-level responses and protected SEO fields that commonly move during releases and client edits. It does not yet claim continuous link-by-link crawling.
See continuous monitoring →Straight answers
Before you check.
The most useful link report distinguishes evidence from uncertainty and states exactly how much it checked.
01What links does RankLatch check?+
The tool reads anchor href values in one page's server-returned HTML, resolves relative addresses, keeps unique same-host HTTP and HTTPS destinations, and checks the first 12 in source order.
02What counts as a broken link?+
RankLatch labels a destination broken only when the final response is HTTP 404, 410, or a 5xx server error. Restricted, rate-limited, unusual, and timed-out responses are separated so uncertainty is not presented as a confirmed failure.
03Are redirecting links broken?+
No. A redirect can work correctly, so RankLatch labels it separately. For permanent redirects, linking directly to the final canonical URL usually reduces unnecessary hops and makes the intended destination clearer.
04Does this crawl the whole website?+
No. It checks one submitted page and at most 12 unique internal destinations. The result reports the total unique links discovered and labels when the live check is capped.
05Does the checker execute JavaScript?+
No. It reads the public server HTML. Links inserted only after browser-side JavaScript runs may not be included.
06Why are external links skipped?+
The bounded free endpoint stays on the submitted hostname to limit abuse, subrequests, and ambiguous third-party blocking. Use a full crawler when every external destination must be tested.
07Can a successful link still be bad for SEO?+
Yes. A 200 response does not prove the destination is relevant, indexable, canonical, internally discoverable elsewhere, or good for users. This checker answers the narrower response-health question.