Free launch tool
Broken Link Checker
Paste a landing page URL and see which links respond, which redirect, and which come back broken. The checker fetches that one page and the links on it. Every row is a real response. Nothing is filled in when a request fails.
What this checker does
Paste the URL of one page, usually the landing page you are about to launch. EarlyHunt fetches that page from its server, reads the anchor tags, and requests each unique http or https link. Relative URLs are resolved against the final page URL, including a base tag when the HTML has one. mailto, tel, javascript, and links that are only #anchors are skipped. The same URL is checked once, and the first anchor text is kept.
The checker does not run JavaScript, click buttons, or submit forms. A link that exists only after a script runs is not in the server HTML, so it is not in the table. The checker also does not crawl onward: it requests a link to read that response, and it does not collect the links on the destination.
How to read a row
OK means the final response was in the 200 range and the URL did not redirect. Redirect means the checker followed a 301, 302, 303, 307, or 308. The final URL is shown so you can see where the link lands. A redirect that ends on a 404 is broken, not a successful redirect.
Broken means a 4xx or 5xx status other than 403 and 429. A 404, a 410, and a 500 are broken. Blocked means the checker refused the address, or the site answered 403 or 429. Those two statuses usually mean an automated client was refused, not that a visitor would see a missing page. Timeout means no status code came back: the host did not resolve, the connection failed, or the time limit was reached. A timeout is not labeled broken.
Limits
At most 150 unique links are checked. If the page has more, the result says so and the rest are left unchecked. A handful of links are requested at a time, each with a short timeout. The page HTML is read only up to 1.5 MB.
Internal means the link host matches the page host. A leading www is ignored, so www.example.com and example.com are the same host. A subdomain such as docs.example.com is external, which keeps a dead docs host visible in the table.
Some sites block this user agent. When they do, the row says blocked. Open that URL in a browser before you treat it as a problem on your landing page.
Before you launch
- Paste the exact URL you will share, including https.
- Fix broken rows first. A 404 or 410 fails for visitors too.
- Open each redirect and confirm the final URL is the page you intended.
- Open blocked and timeout rows in a browser. A 403 or 429 often means the site refused this checker.
- Run the check again after you edit the page. Only the URL you paste is fetched.
When the rows you care about are clean, check the images in the Launch Image Size Checker and pick other directories in the Launch Directory Finder.
Ship the page once the links are clean
EarlyHunt Premium is $19 once if you want a guaranteed dofollow from this site, a skip of the free queue, and Launch Kit graphics. Free launches stay open.
Broken link questions
Does the broken link checker crawl my whole site?
No. It fetches the one URL you paste and checks the links on that page. It does not open those destinations to collect more URLs, and it does not walk your sitemap. A page with more than 150 unique links is capped, and the result says how many were left out.
What is the difference between broken and blocked?
Broken means the destination returned a 4xx or 5xx status, other than 403 and 429. A 404, a 410, and a 500 are broken. Blocked means the checker was refused or would not send the request. A 403 or 429 is blocked, because many sites answer that way to automated clients while the page still loads in a browser. Private, loopback, link-local, and cloud metadata addresses are also blocked and are not requested. A timeout means no status code was received. Timeouts and blocks are not counted as broken.
Will it see links added by JavaScript?
Only if those links are in the HTML the server sends. The checker does not run JavaScript, so a button or a client-rendered menu that never appears as an anchor in the response is omitted. If the table is empty on a page you know is full of links, view the page source and look for anchor tags.
Why are local and metadata addresses refused?
The checker runs on EarlyHunt's server. A link to 127.0.0.1, a private LAN address, a link-local address, or a cloud metadata host such as 169.254.169.254 could point at this server or at cloud credentials. Those URLs are rejected, including when a public hostname resolves to one of them and when a redirect points at one of them. The row is marked blocked and no request is sent.
Do you store the page I submit?
The page is fetched to read its links, and the results are returned to your browser for this check. EarlyHunt does not write the HTML or the result table to the database. Checks are rate limited. Submit a page you are comfortable having this server fetch.