Paste a page's source and SEO Check reads it the way a crawler does. Each finding quotes the tag it came from. A check it could not run is listed as not checked and is never counted as a pass.
The single-page checks from seo-crawl, the open skill that crawls a whole site. Thresholds such as 60 characters for a title are rules of thumb for what search results display, not limits set by any search engine.
A web page is not allowed to read another site's pages, and this tool has no server to fetch them for you. The upside is that it works on pages nobody else can reach: a local build, a staging site, a page behind a login.
Nowhere. It is parsed in your browser as an inert document, so its scripts do not run and nothing in it is loaded or sent.
Then the raw source may be an empty shell, and the tool will say so. Paste the rendered HTML instead: open the browser's developer tools, right-click the <html> element and copy its outer HTML.
No tool can promise that. These are the technical faults that stop a good page being indexed or shown properly. Whether it ranks depends on the content, the competition and links, none of which is in the source.
Anything the server sends rather than the page: status codes, redirects, headers, robots.txt, the sitemap. It also has no rankings, clicks or backlinks, and it does not measure speed. The result lists these as not checked.