Broken Link Checker – Find 404s & Dead Links on Any Website

Broken Link Checker crawls your website, checks every link it finds, and gives you a list of dead links, 404 pages, broken images and failing redirects. For each one it shows the page and anchor text where the link appears, so you know exactly what to fix. It uses plain HTTP and no browser, so it is fast and cheap: $0.30 per 1,000 links checked.

What it does

• Crawls your site from one or more start URLs or a sitemap.xml. It stays on your domain, goes breadth-first, and you can limit it with maxPages, maxDepth and include/exclude globs.
• Extracts every link: <a href>, <area>, images (<img src>, srcset, <picture>), and optionally scripts, stylesheets, icons, iframes and video/audio sources.
• Checks each unique URL once. It sends HEAD first and falls back to GET when a server rejects HEAD. Redirects are followed by hand, so every hop of the redirect chain is recorded.
• Finds every kind of broken link: 4xx (404, 410, 403…), 5xx, DNS errors (dead domains), timeouts, SSL/certificate errors, connection refused/reset, redirect loops, too many redirects and invalid redirects.
• Shows where to fix it: foundOn lists each page that contains the link, with its anchor text (or image alt text) and HTML element.
• Flags likely false positives. External sites that block bots (403, 429, LinkedIn's 999) are marked possibleFalsePositive and are not mixed in with real dead links.
• Writes a SUMMARY to the key-value store: pages crawled, links checked, broken counts by status code and by link type, and the top 20 pages with the most broken links.
• Is polite: it respects robots.txt (Crawl-delay included), throttles requests per host, and sends an honest User-Agent with a contact address.

Use cases

• Website audit before launch or after a migration: catch 404s, broken images and redirect chains.
• Scheduled monitoring: run it weekly with onlyBroken: true and get an alert or webhook only when new dead links appear.
• SEO cleanup: broken internal links waste crawl budget and link equity. Fix them using the "Where to fix" view.
• Content teams and agencies: find dead outbound links in blogs, docs and knowledge bases, and in client sites.
• Sitemap health: feed in sitemap.xml and see which listed URLs are broken.

Input example

{
  "startUrls": [{ "url": "https://www.example.com/" }],
  "maxPages": 500,
  "maxDepth": 20,
  "checkExternalLinks": true,
  "checkImages": true,
  "checkScriptsAndStyles": false,
  "excludeGlobs": [{ "glob": "https://www.example.com/cart/**" }],
  "onlyBroken": true,
  "concurrency": 5,
  "timeoutSecs": 15,
  "respectRobotsTxt": true
}

If you leave the input empty, the Actor runs a 45-second demo on crawler-test.com, a public crawler test site full of intentionally broken links.

Field	Default	Meaning
startUrls	–	Website home page(s) or sitemap URL(s)
maxPages	100	HTML pages scanned for links, per site
maxDepth	20	Clicks from the start page (0 = start page only)
checkExternalLinks	true	Also check links to other domains (never crawled)
checkImages	true	Check <img>, srcset, <picture>
checkScriptsAndStyles	false	Check scripts, CSS, icons, iframes, media
onlyBroken	true	Output only broken links, or every checked link
includeGlobs / excludeGlobs	–	Limit which pages are crawled / skip URLs entirely
maxLinksToCheck	10,000	Cost cap: unique links checked per run
useSitemap	false	Also crawl pages listed in robots.txt sitemaps
respectRobotsTxt	true	Obey robots.txt and Crawl-delay
concurrency	5	Parallel requests to your site
timeoutSecs	15	A link slower than this is reported as TIMEOUT

Output example

One dataset row per broken link (or per checked link when onlyBroken is off):

{
  "url": "https://crawler-test.com/redirects/invalid_redirect",
  "isBroken": true,
  "status": 404,
  "statusText": "Not Found",
  "errorType": "HTTP_4XX",
  "error": "HTTP 404 Not Found",
  "possibleFalsePositive": false,
  "redirectChain": [{ "url": "https://crawler-test.com/redirects/invalid_redirect", "status": 303 }],
  "redirectCount": 1,
  "finalUrl": "https://crawler-test.com/images/foo.jpg",
  "isExternal": false,
  "linkType": "page",
  "foundOnCount": 1,
  "foundOn": [{ "pageUrl": "https://crawler-test.com/", "anchorText": "Redirect 303 To 404", "element": "a" }],
  "firstSeenAt": "2026-10-11T12:00:48Z",
  "checkedAt": "2026-10-11T12:00:59Z",
  "responseTimeMs": 919,
  "method": "HEAD"
}

errorType is one of HTTP_4XX, HTTP_5XX, DNS_ERROR, TIMEOUT, SSL_ERROR, CONNECTION_REFUSED, CONNECTION_RESET, CONNECTION_ERROR, PROTOCOL_ERROR, REDIRECT_LOOP, TOO_MANY_REDIRECTS, BAD_REDIRECT.

The SUMMARY record in the key-value store looks like this:

{
  "pagesCrawled": 10, "linksChecked": 394, "brokenLinks": 72, "brokenConfirmed": 70, "possibleFalsePositives": 2,
  "brokenInternal": 70, "brokenExternal": 2,
  "brokenByStatus": { "404": 8, "403": 3, "DNS_ERROR": 2, "REDIRECT_LOOP": 2, "TIMEOUT": 2, "500": 2 },
  "brokenByLinkType": { "page": 72 },
  "topPagesWithBrokenLinks": [{ "pageUrl": "https://crawler-test.com/", "brokenLinks": 72 }],
  "notChecked": { "robotsTxtDisallowed": 17 }
}

The Output tab has three views: Broken links, Where to fix (one row per page + anchor text, ready to hand to an editor) and Redirect chains.

Pricing

Pay per event: $0.30 per 1,000 unique links checked, plus the Apify start fee of $0.00005. Crawled pages count as links. Duplicates, links skipped by robots.txt or your filters, and the summary are free. Set a maximum charge on the run and the Actor stops cleanly at that limit, keeping all results collected so far.

Actor	Users (30 days)	Price
This Actor	new	$0.30 per 1,000 links checked
khadinakbar/broken-link-checker	31	$1.00 per 1,000 links checked
parseforge/broken-link-checker	7 (rating 2.7)	see Store
logiover broken link checker	3	see Store
santamaria-automations broken link checker	2	see Store

Typical costs: a 100-page site with about 1,200 links costs $0.36. The empty-input demo (394 links) costs $0.12.

FAQ

Does it run JavaScript? No. It reads the HTML the server returns, which covers most websites, including WordPress, Shopify, Webflow, docs sites and static sites. Links that only appear after client-side rendering (some single-page apps) are not seen.

Why is a link that works in my browser marked 403? Some sites (LinkedIn, npm, some CDNs) block automated requests. These links are flagged possibleFalsePositive: true with a note, sorted last, and counted separately in SUMMARY.possibleFalsePositives.

Does it overload my server? No. By default it sends at most 5 parallel requests to your site and honors robots.txt Crawl-delay (then 1 request at a time). External hosts get at most 2 parallel requests. Each URL is requested only once.

Can I check only one page? Yes. Set maxDepth: 0: the start page is scanned and all of its links are checked.

Can I check a list of URLs without crawling? Use our Bulk URL Status Checker for that. This Actor is built for finding links inside pages.

What is the User-Agent? Mozilla/5.0 (compatible; BrokenLinkChecker/0.1; +https://apify.com/gazidev/broken-link-checker; iletisim.ortalamabiri@gmail.com). robots.txt rules for BrokenLinkChecker (or *) are applied.

What if a run is huge? maxPages and maxLinksToCheck cap the work, and the run's maximum charge caps the cost. When the run timeout is near, the Actor stops early and still saves everything it has.

Use with AI agents / Apify MCP

This Actor works as a tool for AI agents through the Apify MCP server (Claude, ChatGPT, Cursor, and so on). Ask your agent to "check https://example.com for broken links". It calls gazidev/broken-link-checker with startUrls and gets back compact JSON rows: url, status, errorType and foundOn with page and anchor text. The agent can then summarize them or open fix tickets. Use maxPages and maxLinksToCheck to keep agent runs small and predictable. The same input works from the Apify API, Python/JS clients, Make, Zapier and n8n.