// Collects distinct, already-resolved-absolute <img> source URLs from the
// rendered page. `currentSrc` (falling back to `src`) is a browser-resolved
// absolute URL — no `<base href>` handling needed here, unlike raw attribute
// values elsewhere in this codebase's crawler.
//
// Bounded client-side (MAX_URLS) purely to keep the evaluate() payload small;
// the real per-page fetch cap (MAX_IMAGES_CHECKED) is enforced in Rust.
const MAX_URLS = 200;
const urls = ;
const seen = ;
return ;