Response time is within limits. A crawler that waits two seconds for the first byte may read half the page or give up.
The check measures time to first byte on each priority page. Fetchers run with budgets; a slow first byte pushes a fetch toward a timeout and a truncated read.
A crawler that waits two seconds for the first byte may read half the page or give up. Partial fetches mean partial understanding — the opening gets read, the facts further down do not.
curl -o /dev/null -w "%{time_starttransfer}" your page from a few locations. Under half a second is comfortable; over a second and a half is where fetchers start truncating.
What the agent does: speed lives in your hosting; the agent reports the number per page and what it points at.
Part of the 23 checks · group: Page content — text without JavaScript, response time, internal links, answer-first structure, self-contained sections, headings/dates/authorship.
It is the first of them. Core Web Vitals measure the experience in a browser; a crawler only waits for bytes, so time to first byte is the one that matters here.
The page check reads your public pages from outside and runs all 22 — free, no signup, no AI engine involved.