One page per job, each with a request you can paste. Everything here works on the free tier.
Capture a page as PDF or PNG with the upstream status, content type and render time recorded alongside it, so an archive carries the conditions it was taken under.
Show a thumbnail and a summary for any link a user pastes, from one render, without parsing Open Graph tags that half the web does not ship.
Capture a competitor's pricing page or a supplier's terms on a schedule, diff the Markdown to know what changed, and keep the image as proof.
Generate the social preview image for a page by screenshotting the page itself, or a template you control, with one GET request that fits inside a meta tag.
Invoices, reports and certificates rendered from a URL you control, with the settings that matter when the HTML is yours rather than somebody else's page.
Two tools over MCP, screenshot and read_page, so an agent can see a page's layout or read its text without you running a browser inside your agent process.
A column of URLs becomes a column of thumbnails with one IMAGE formula, because the API returns image bytes rather than JSON.
Capture the same pages on every deploy and compare them against a baseline. No Puppeteer in your pipeline, no Chromium download, no container that needs fonts.