Web hosting
Website broken links: find 404s and decide what should replace them
Broken links can be internal routing defects, deleted assets, stale external references, malformed relative URLs, permission errors, or JavaScript-generated paths. A useful link audit distinguishes ownership and status so you do not “fix” a temporary external outage by deleting valuable context.
Crawl rendered and source-visible links
Check anchors, images, scripts, stylesheets, canonical links, sitemap entries, and other static references. For JavaScript-heavy interactions, browser tests may be needed beyond an HTML parser.
Normalize fragments/query strings carefully so distinct application routes are not collapsed incorrectly.
Classify internal failures first
Internal 404/500/redirect loops are under your control and can break navigation or assets. Decide whether the correct action is restore the page, update the link, or add a relevant redirect from an old public route.
Do not redirect every missing URL to the homepage; that hides errors and creates poor search/user behavior.
Treat external links as monitored references
A vendor/documentation URL can fail transiently or block automated requests while still working for users. Record status and manually verify important references before removing them.
Replace obsolete external documentation with a current authoritative source where available.
Prevent regression in the build
Run link/static-asset validation on generated output in CI and browser-test route samples. Keep a deliberate allowlist only for known external contracts—not as a way to ignore internal failures.
- Check internal links/assets first.
- Use specific redirects for moved pages.
- Manually verify important external failures.
- Fail CI on broken internal references.