Broken Link Checker: Find Broken Links, Images & Redirects

What does Broken Link Checker do?

Broken Link Checker crawls your website and checks every link, image, script, stylesheet and embed once, then tells you exactly which links are broken and on which pages they appear. Enter a website such as example.com and it finds 404 pages, server errors, dead domains (DNS failures), timeouts, SSL problems, redirect loops, long redirect chains and insecure (mixed content) resources.

Each problem is one row with every page that links to it, the link text, a plain-language explanation and the full redirect chain, so you know what to fix and where. Because it runs on the Apify platform, you get API access, scheduling (weekly site-health checks), webhooks, integrations (Make, Zapier, n8n, Google Sheets, Slack) and run monitoring.

What you pay for

Every unique link checked is charged once, including links that turn out to be fine and links that are filtered out of the results when Report all links is off. A link that appears on 50 pages is checked and charged once. Crawling the pages to find those links is free.

Two ceilings keep a run predictable: maximum pages to crawl (default 500) and maximum links to check (default 5,000) per website. Sites with faceted search or calendars can generate far more links than pages, so raise the link ceiling deliberately rather than by accident.

You are never charged for a failed or empty result

This Actor charges per link checked, because the check is the work. Its item-level status field is the HTTP status code; the result quality lives in category.

category	Meaning	Charged
ok	The link works	Yes — every link checked is work done
broken	404, dead domain, timeout or TLS failure	Yes — the check is the work
redirect / possibly-blocked	Moved, or the site refused automated checks	Yes — the check is the work
(link never checked)	The website could not be crawled at all	No
(stopped at your cost limit)	Remaining links are not checked	No

If every input fails, the run itself fails with an explanation instead of reporting success with no data.

Why use Broken Link Checker?

• SEO and user experience: broken links waste crawl budget and frustrate visitors; redirect chains slow pages down.
• Website migrations and redesigns: verify that nothing broke after moving pages or changing URLs.
• Content audits: find dead outbound links in old blog posts and documentation.
• Fewer false alarms: sites that block bots (HTTP 401/403/429, Cloudflare challenges) are marked possibly-blocked, not broken, and every failed HEAD request is confirmed with a GET before it is reported.
• Scheduled monitoring: run it weekly and get a webhook or email only when new broken links appear.

How to find broken links on a website

1. Click Try for free.
2. In Websites to check, enter one or more websites or pages (for example example.com).
3. Optionally set Maximum pages to crawl per website and Maximum crawl depth.
4. Click Start.
5. Open the Output tab. Overview lists broken, blocked and redirected links with the pages where they appear; Redirects shows each redirect chain. A summary per website is in the key-value store as SITE_SUMMARY.

Input

Field	Description	Default
Websites to check	Websites, domains or start pages. Pages on the same website are crawled; links to other websites are checked but not crawled	–
Maximum pages to crawl per website	Crawl limit (up to 2,000); all links on crawled pages are still checked	500
Maximum links to check per website	Cost ceiling: the run stops checking a website after this many unique links	5,000
Maximum crawl depth	Clicks away from the start page (0 = only the start page's links)	5
Check links to other websites	Check external links too	on
Check images, scripts and stylesheets	Also check img (including srcset), script, stylesheets, icons, iframe, video, audio	on
Report redirected links	Include working links that redirect, with the chain	on
Report all links	Save working links as well	off
Detect soft 404 pages	Pages that return 200 but say "not found" in the title count as broken	off
Also crawl pages from the sitemap	Add sitemap pages, so orphan pages are checked	off
Skip URLs matching	Glob (*/tag/*), text (/cart) or regex between slashes	none
Respect robots.txt	Skip pages and links the website's robots.txt disallows. Turn off only for websites you own	on
Parallel requests	Total parallel checks (same-host requests are limited to half, 4–10)	10
Request timeout (seconds)	Slower links are reported as timeout	15
Identify as a web browser	Send a Chrome User-Agent instead of BrokenLinkChecker/1.0	off
User agent	Send your own User-Agent string instead of the built-in one. Overrides the toggle above.	–
Report links slower than (ms)	Mark links that took at least this long with isSlow. 0 turns it off.	0 (off)
Maximum redirects to follow	1–20. A link needing more hops is reported as too-many-redirects, which counts as broken.	10

Example input:

{
  "startUrls": ["example.com"],
  "maxPagesPerSite": 1000,
  "excludePatterns": ["*/tag/*", "/\\?replytocom=/"]
}

Output

One item per problem link. Real examples from a run on 15 September 2026 against crawler-test.com (a public test site full of intentional errors):

{
  "site": "https://crawler-test.com",
  "linkUrl": "https://crawler-test.com/redirects/redirect_1",
  "finalUrl": "https://crawler-test.com/redirects/redirect_target",
  "status": 200,
  "statusCode": 200,
  "category": "redirect",
  "message": null,
  "isExternal": false,
  "linkType": "a",
  "mixedContent": false,
  "redirectChain": [
    { "url": "https://crawler-test.com/redirects/redirect_1", "status": 301 },
    { "url": "https://crawler-test.com/redirects/redirect_target", "status": 200 }
  ],
  "sourceCount": 1,
  "sourcePages": [
    { "pageUrl": "https://crawler-test.com/", "anchorText": "Redirect 301", "element": "a", "attribute": "href" }
  ],
  "checkedAt": "2026-09-15T12:07:24.193340+00:00"
}

A broken link looks like "status": 404, "category": "broken", "message": "HTTP 404: the page was not found."; a dead domain like "status": "dns-error", "message": "The domain name could not be resolved (DNS failure); the domain may not exist.".

That run crawled 29 pages and checked 414 unique links in 60 seconds: 70 broken, 11 redirects, 5 possibly blocked, 2 mixed-content resources, 13 URLs skipped because robots.txt disallows them.

You can download the dataset in various formats such as JSON, HTML, CSV, or Excel.

Output fields

Field	Meaning
category	broken, redirect, possibly-blocked or ok
status	HTTP status code, or an error: dns-error, timeout, ssl-error, connection-refused, connection-error, too-many-redirects, invalid-redirect, protocol-error, soft-404
statusCode, message	Numeric HTTP status (when there was a response) and a plain-language explanation
linkUrl, finalUrl	Link as found (normalized, without #fragment) and where it ends up after redirects
redirectChain	Every hop with its status code
linkType	Element the link came from: a, img, script, link, iframe, source, video, audio, sitemap
isExternal, mixedContent	Points to another website; http:// resource embedded in an https:// page
sourcePages, sourceCount	Up to 20 pages containing the link (with anchor text) and the total number of occurrences
site, checkedAt	Website crawled and check time (UTC)

SITE_SUMMARY (key-value store) has per website: pages crawled, links checked, broken, redirects, blocked, mixed content, URLs skipped by robots.txt, whether robots.txt was respected (robotsTxtRespected), and the 10 pages with the most broken links.

How much does it cost to check a website for broken links?

This Actor uses pay-per-event pricing: you pay per unique link checked (each URL is checked once per run, however many pages link to it). Crawled pages count as links. Platform compute is included. See the Pricing tab for the price per 1,000 links. A typical 200-page small-business site has 1,000–3,000 unique links.

You can set a maximum cost per run: the Actor stops starting new checks when the budget is reached and still saves everything checked so far.

Finding slow links as well as broken ones

Set Report links slower than (ms) and every checked link gains two fields: responseTimeMs, the wall time for the whole check with redirect hops included and our own polite pacing excluded, and isSlow.

A slow link is never counted as broken. It keeps whatever verdict its status code earns, and isSlow sits beside that verdict rather than replacing it — a page that takes four seconds and returns 200 is a working page that needs attention, not a dead one, and reporting it as dead would put a false claim in front of whoever reads your report.

responseTimeMs is filled in for every check, even without a threshold, so you can sort by it afterwards. It is null only when no response arrived at all, such as a DNS failure — an unknown time is reported as unknown rather than guessed at as zero.

Tips

• Start small: try 50 pages first to see how many links your site has.
• Big sites: raise Parallel requests to 20–30 and exclude endless URL patterns (calendars, filters, ?sort=), or turn off external links.
• Many possibly-blocked links? Social networks and some CDNs block automated checks. Turn on Identify as a web browser, or simply verify those few links manually.
• Long crawls: increase the run timeout; the Actor stops starting new checks two minutes before the timeout and saves results.
• Orphan pages: turn on Also crawl pages from the sitemap.

Related tools

More website audit tools:

• Sitemap URL Extractor: list every URL of a website from its sitemaps, or only new and removed pages.
• SEO Audit & Metadata Checker: check titles, meta descriptions, canonicals and other on-page SEO for pages or whole sites.
• Website Tech Stack Detector: detect the CMS, frameworks, analytics and hosting of any website.

FAQ, limitations and support

Does it respect robots.txt? Yes, by default. Pages disallowed for crawlers are not requested and are counted in SITE_SUMMARY.

Can I check my own site even when its robots.txt blocks crawlers? Yes — turn Respect robots.txt off. Many sites Disallow: / for everything except search engines, which stops a link check of your own site before it starts. With the toggle off we do not request /robots.txt at all: nothing is skipped for robots reasons, skippedByRobotsTxt and skippedUnreadableRobotsTxt are 0, robotsTxtRespected is false in SITE_SUMMARY, and the run's status message says so. Because robots.txt is not read, sitemaps are then looked for at the usual addresses (/sitemap.xml, /sitemap_index.xml, …) instead of the ones robots.txt names. Only turn it off for websites you own or have permission to crawl — you take responsibility for the requests, and every other limit still applies unchanged: the per-website link and page ceilings, the crawl depth, the parallel-request limits and the run timeout.

What happens when a site won't let us read its robots.txt? Some sites put their /robots.txt behind bot protection and answer 403 or time out. We then can't tell what the site allows, so we follow one rule everywhere: the address you entered is still read, and addresses we would have worked out ourselves are not. A site with no robots.txt at all (404) is treated as allowing everything. For this Actor that means the start URL you entered is still checked, but the crawl does not follow links deeper into that site and no sitemap is guessed. SITE_SUMMARY counts those separately as skippedUnreadableRobotsTxt, so you can tell them from pages a real Disallow rule covered. If it is your own site, turning Respect robots.txt off lifts this too. It uses at most 4–10 parallel requests per host with small delays.

Does it run JavaScript? No. It reads the HTML the server sends, like search engines' first pass. Links added only by JavaScript are not found.

What about email and phone links? mailto:, tel: and javascript: links are skipped and never saved.

Why is a working link marked possibly-blocked? The server refused an automated request (401, 403, 429, 999 or a Cloudflare/Akamai challenge). It usually works in a browser.

Something wrong or missing? Open an issue in the Issues tab. Custom website audits and integrations are available on request.

Only check websites you are allowed to crawl.