Broken Link Checker

A fast, no-browser broken link checker built for teams, not just one-off checks. Point it at one or more pages; it extracts every link, checks them in parallel, and hands back a per-page letter grade plus a triage-ready, severity-tagged list -- so you catch dead links before your users, your boss, or Google does.

What you get

Per page

• A letter grade (A+ to F) and 0-100 score summarizing link health at a glance
• A one-line executive summary ("3 of 42 links are broken -- review recommended")

Per link

• Status code and pass/fail, checked concurrently (configurable, up to 20 at once) so a 200-link page finishes in seconds, not minutes
• A severity tag -- critical (dead link), warning (redirects somewhere else), or ok -- for instant triage in any BI tool or spreadsheet
• Redirect detection with the final destination URL, and response time in milliseconds
• Anchor text, so you know exactly which piece of link copy points where
• Internal vs. external flag, with an option to skip external links entirely

Why this instead of checking manually

	Manual spot-checks	Broken Link Checker
Coverage	Whatever you click	Every link on the page, automatically
Speed	Minutes per page	Seconds, via parallel checking
Triage	You guess what's urgent	Pre-classified severity + letter grade
Repeatability	Forgotten next sprint	One click, or schedule it
Output	Notes in a doc	Structured dataset, exportable to CSV/JSON/BI tools

Who this is for

• Dev and QA teams running a pre-launch or post-migration link audit
• Content and marketing teams auditing old blog posts or resource pages
• Agencies checking a client's site health alongside an SEO audit
• Anyone wiring a scheduled link-rot regression check into their workflow

How to use it

1. Add one or more pages to scan. Each page's HTML is fetched once and every link on it is checked.
2. Leave check external links on (default) to check every link, or turn it off to only check links on the same domain.
3. Tune concurrency (default 8) to trade speed for gentleness on slow servers, and max links per page to bound very link-heavy pages.
4. Run it once for a snapshot, or schedule it to catch link rot as it happens.

Example output (one item)

{
  "sourceUrl": "https://example.com/",
  "linkUrl": "https://example.com/old-page",
  "linkText": "Read more",
  "isExternal": false,
  "statusCode": 404,
  "ok": false,
  "severity": "critical",
  "redirected": false,
  "finalUrl": null,
  "responseTimeMs": 412,
  "error": null,
  "pageTotalLinksChecked": 42,
  "pageBrokenLinks": 3,
  "pageScore": 93,
  "pageGrade": "A-",
  "pageSummary": "3 of 42 link(s) are broken -- review recommended.",
  "checkedAt": "2026-08-20T12:00:00.000Z"
}

Pricing

This Actor uses pay-per-event pricing:

• Page scanned -- charged once per page URL processed (covers fetching the page and extracting its links), whether the page loaded successfully or not.
• Link checked -- charged once per unique link actually checked on a page.

You only pay for links that get checked -- if a page has fewer links than your max, or you skip external links, you pay less. Parallel checking doesn't cost more -- it just finishes faster.

FAQ

Does this render JavaScript? No -- it reads the raw HTML response and extracts <a href> links from it, the same way a search engine's initial crawl would see the page. Links injected purely by client-side JavaScript won't be found.

Does it follow links and crawl deeper into the site? No -- this checks the links found on the pages you give it, one level deep. It's a fast link-rot check, not a full-site crawler.

What counts as "broken" vs. a "warning"? critical means the request failed outright (non-2xx status after redirects, timeout, DNS error, connection reset). warning means the link resolved successfully but redirected somewhere else -- not broken, but often worth cleaning up (e.g. pointing links straight at the final URL).

Will higher concurrency get me rate-limited? It can, on servers with aggressive rate limiting. Start at the default (8) and lower it if you see a spike in timeouts for a particular site.

Does this send any data anywhere besides your own Apify account? No. It only makes standard HTTP requests to the pages and links you provide and returns what it finds -- no login, no third-party analytics calls.