Test pages for how Google's renderer loads a page, made for GSC Full Fetch & Render. Run one in Search Console's URL Inspection, live or crawled: each page writes what it saw into its own rendered HTML, so the HTML tab of the tested page shows the result.
wrs/: how far the viewport reaches down the page and across it, logged as it changes, and which loading="lazy" images Googlebot requests, down the page and along a carousel. ?h= sets how tall the page is.
wrs/static.html: the same reading on a static page, over 10,000 px of text all in its HTML and nothing built by script: innerHeight read once at the top, as the parser reaches it, then every resize.
wrs/2mb.html: a 3 MB HTML file with markers of their byte offset, to see how much of it Googlebot reads and URL Inspection shows.
wrs/size.html: a rendered HTML of a set number of paragraphs and kilobytes, of random or repeated words, to see what decides where the HTML tab stops.
wrs/nodes.html: which DOM nodes count toward that point.
wrs/nodes-index.html: a static page of 150,000 DOM nodes with made-up words at known nodes, to see which of them Google indexes.
wrs/shadow.html: what the rendered HTML keeps of open, closed and declarative shadow roots, of the page's text slotted into one or left out of it, and of iframes of an indexable page and of a noindex one, each with a made-up word to see whether Google indexes it as part of the page.
Source: github.com/natzir/gsc-full-render.