Broken Link Checker
Check every link on a page and see which are broken, blocked or redirected.
Check the links of a page
Fetching the page from MySmartCoPilot’s server…
Links
| What happened |
|---|
About the Broken Link Checker
Enter a page address and the checker fetches the page, lists its links — and, if you ask, its images, scripts, stylesheets and embeds — and requests each unique address to see what happens. It uses HEAD first and falls back to GET when a server refuses HEAD, follows up to five redirects, and shows every link as Broken (404, 410, a domain that does not exist, a redirect loop), Could not verify (the site refuses automated requests, times out or has a server error), Redirected (with the final address) or Working, together with the anchor text, the line in the page and how many times the link appears.
Links that only point into the page (#section) are checked against the ids on the page, and mailto:, tel: and javascript: links are listed but not requested. Results can be filtered, sorted and exported as CSV.
How to use it
- Enter the page address in Page URL, or switch to Paste HTML and paste the page source with its address, so relative links can be turned into full addresses.
- Choose which links to check: all, external only (links to other websites) or internal only, and tick Also check images, scripts, stylesheets and embeds to include those.
- Press Check links. The table fills in as groups of ten links are checked; you can cancel at any time and keep what was checked.
- Start with Broken. Each row shows the status, the link, its anchor text and the line in the page, so you can find and fix it. Open Could not verify links in a browser to be sure.
- Fix redirected internal links by pointing them at the final address. Copy the report or download the table as CSV to hand over or compare after the fix.
Examples
https://www.example.com/blog/
Broken 404 https://www.example.com/old-guide/ “our guide to chai” line 128 Not found (404): there is no page at this address.
The checker confirms a 404 with a GET request before it calls a link broken, because some servers answer HEAD requests wrongly.
https://www.linkedin.com/company/example
Could not verify 403 The server refuses MySmartCoPilot’s request (403 Forbidden). Many sites block automated checks: open the link in a browser to be sure.
Login walls and bot protection answer 401, 403, 429 or non-standard codes such as 999 to tools, while the same link opens in a browser. These are not reported as broken.
Common uses
- Cleaning up a blog post or landing page before a launch or newsletter: dead links, moved pages and missing images.
- Finding internal links that redirect (after a site migration or a change of URL structure) so they can point at the final address.
- Checking a page of references or resources, which goes stale faster than any other kind of page.
- Finding broken in-page anchors and links to domains that have expired.
What counts as broken
- Broken: the server says the page does not exist (
404 Not Found,410 Gone, RFC 9110 §15.5), the domain name does not exist (a failed DNS lookup), the redirects form a loop or are longer than five hops. - Could not verify: other 4xx answers (401 login needed, 403 forbidden, 429 too many requests, 451, non-standard codes such as 999), timeouts, failed connections and 5xx server errors. They often mean that a site blocks automated requests or has a temporary problem, and the link may work in a browser.
- Redirected: the link ends on a working page after one or more redirects; the final address is shown, and a permanent (301/308) or temporary (302/303/307) redirect is noted. Google treats every 4xx except 429 as “this content does not exist” for search; the checker is stricter about what it calls broken, because a link checker only sees what the site says to MySmartCoPilot’s server.
- Not checked: private and internal addresses, non-standard ports, and anything left when a run is cancelled or reaches a limit.
What the checker sends, and its limits
Requests come from Cloudflare’s network with the user agent MySmartCoPilotBot/1.0 and no cookies. Only http:// and https:// on ports 80 and 443 are requested; private, loopback, link-local and cloud-metadata addresses are never fetched, each new host is looked up over HTTPS first, every redirect target is checked again, and each request times out after 8 seconds. A website is asked about one link at a time and only a limited number of times in a few minutes, so the checker cannot be used to overload it.
Free runs check up to 200 unique addresses per page. A page with 200 links needs about 21 server requests (one for the page and one for every 10 links); your hourly allowance of requests is shared by all MySmartCoPilot checks and the page tells you when it is running low. For very large pages, check external and internal links in separate runs.
Limitations
- Only the first 200 unique addresses of a page are checked in one run (choose external-only or internal-only to see the rest). Site-wide crawling is not available.
- Soft 404s — pages that answer 200 with a “not found” message — cannot be detected, and neither can links that only work when you are logged in.
- The checker reads the HTML as the server sends it and never runs scripts, so links added by JavaScript after the page loads are not found.
- A site can answer MySmartCoPilot’s server differently from your browser (bot protection, country rules). Links in the “Could not verify” group deserve a manual look.
- A link to a page may work while an anchor in it (
page.html#section) does not exist: only anchors inside the checked page are verified. - Each visitor has an hourly allowance of server requests, and the whole site a daily one, to stay within free hosting limits. If the allowance runs out, the run stops and keeps what was checked.
Privacy
The page address and then the addresses of its links are sent to MySmartCoPilot’s server, which requests them from their websites. MySmartCoPilot does not store the addresses or the results; the server log records only the host name, status code and time of each request. In Paste HTML mode the HTML stays in your browser and only the link addresses are sent.
Frequently asked questions
Why is a link that works in my browser reported as “Could not verify”?
Many websites answer automated requests with 403, 429 or a non-standard code, or need cookies or a login, while browsers get the page. The checker only sees what the site tells MySmartCoPilot’s server, so it does not call those links broken. Open them in your browser; if they work there, they are fine for visitors.
Does a redirected link count as broken?
No. A redirected link works, but every redirect costs a request, so it is better to link to the final address, especially for your own pages. The checker shows the final address and whether the redirect is permanent (301, 308) or temporary.
Are broken links bad for SEO?
Google’s link guidance (link best practices) is about making links crawlable and anchor text descriptive, and it says every page you care about should be linked from another page of your site; it does not describe a penalty for broken links. They still send visitors and crawlers to dead ends, and a broken internal link can cut off the pages behind it, so fix those first.
Why do some images and scripts show as broken?
When you tick the option to check images, scripts and stylesheets, every file the page loads is requested too. A 404 there means the page is missing a picture or a script. Files on servers that block hotlinking may show as “Could not verify”.
Can I check a whole site?
Not in one run: the checker reads one page at a time. To cover a site, check its important pages one by one (the sitemap lists them, and the Sitemap Validator can test the addresses in it).