Check Live Site for Broken Links
Crawls the links on your built site and flags every one that returns an error.
Anyone about to launch or relaunch a site who wants to know which links are dead before a visitor finds out.
The problem
A site can accumulate broken links over time as pages move or get deleted, and there is no way to notice this by clicking around yourself. Checking every link by hand does not scale past a handful of pages.
1. Work list
The command that lists every item. No AI involved.
grep -orhE 'href="https?://[^"]+"' ./dist | sed -E 's/href="([^"]+)"/\1/' | sort -u2. Unit of work
One item is one URL found in an href attribute.
3. The check
The checker runs curl -s -o /dev/null -w '%{http_code}' -L --max-time 10 <url>. Exit code 0 if the status code is under 400. Exit code 1 if the status code is 400 or higher, or the request times out, printing the status code back to the AI.
4. Stopping rule
Each broken link gets up to 3 fix attempts (updating or removing it) before quarantine, ending when every URL has been checked.
What this cannot catch
A 200 status code only means the server responded. It does not prove the page still has the content the link was supposed to point to.
Build it
Install justloopit, then describe this task in your AI coding tool. It will ask about your project, then write the checker, the runner, and the spec.