Broken Link Checker

Every link rots eventually — a page gets renamed, a domain gets retired during a migration, or someone fat-fingers a URL in a CMS field and it ships to production untested. Broken Link Checker audits a pasted link inventory (a simple source,target CSV, the kind most CMS and static-site tools can export) and flags the usual suspects: hosts that are not on your known-good list, internal links still using insecure http://, and links whose target looks malformed or unexpectedly external.

It deliberately does not crawl the web. A browser tab cannot safely fetch arbitrary third-party URLs to check status codes, and doing so from a client-side tool would mean either sending your entire link inventory to a server or hitting rate limits and CORS walls on every request. Instead it reasons about the parts of a URL that are cheap and reliable to inspect — the host, the scheme, and whether the path is absolute, root-relative, or protocol-relative — which is exactly where the most common and most consequential broken-link problems actually live: a domain move that leaves old absolute links behind, an http:// link that slipped past a find-and-replace, or a typo host that nobody will notice until a customer clicks it.

• Internal vs external classification — by comparing target host against source host, with "www." normalized away on both sides.
• Unlisted-host flags — paste your canonical/trusted hosts and anything else in the inventory is called out by name with a count.
• Insecure link detection — http:// targets are tallied separately, with special attention to ones that are otherwise internal (mixed content).
• Relative and protocol-relative buckets — /path and //host/path links are counted apart from absolute URLs, since they behave differently across environments.
• Distinct external domains — a ranked list of every external host referenced, and how many times, so a link audit doubles as a third-party dependency inventory.
• Other-scheme handling — mailto:, tel:, and similar links are recognized and set aside instead of miscounted.

1. Paste your CSV rows in source,target form — export them from a CMS, sitemap crawler, or spreadsheet.
2. Add your canonical hosts, one per line, in the optional second box (skip this if you only want the internal/external/relative breakdown).
3. Audit. Stat tiles and a full Markdown report with per-bucket counts and a flagged-host list appear immediately.
4. Fix anything unlisted or insecure first — those are the entries most likely to be an actual mistake rather than an intentional external reference.

Example 1 — CMS export before a crawl. Export every hyperlink from your CMS as a CSV, run the audit, and clean up the unlisted hosts before an SEO crawler or Search Console finds them and dings your site for broken outbound links.

Example 2 — Post-migration cleanup. After moving from www.old-brand.com to example.com, paste the inventory and use the canonical-hosts box to confirm nothing still resolves to the retired www host.

Example 3 — Mixed-content sweep. Before enforcing HSTS or a strict Content-Security-Policy, run the audit to generate a list of every internal http:// link that would suddenly break or get silently upgraded.

Example 4 — Third-party dependency review. The distinct-external-domains list doubles as a quick inventory of every outside service your pages link to — useful when doing a privacy or vendor review.

• Prevent 404s before launch — catch bad or unlisted hosts while the fix is still a find-and-replace, not a customer complaint.
• Find mixed content early — insecure internal links surface before a browser starts blocking or warning about them.
• Audits at CSV scale — paste thousands of rows and get sorted, counted buckets instead of scrolling a spreadsheet by hand.
• No crawler required — works from an exported inventory, so there is nothing to rate-limit, authenticate, or get blocked by robots.txt.
• Private — the entire audit runs in your browser; the link list is never uploaded anywhere.