Web & SEO Utilities

New: Website Screenshot & Browser Inspector — see a page at every size, and why it loads the way it does

Capture a public page in a sandboxed Chromium at up to three screen sizes, full page or visible area, as PNG, JPEG or WebP — with load timings, every request in a waterfall, redirects, headers and TLS.

By NetTools.tech · · 4 min read

A redesign went live and someone says the mobile menu covers the content. A landing page looks fine on your laptop, but colleagues see a consent wall first. A page suddenly takes seconds to show anything, and you need to know which requests are responsible. In each case you want to see the page from outside your own browser, cache and network.

The new Website Screenshot & Browser Inspector does that. It opens a public page in a sandboxed Chromium on the NetTools server, captures it at up to three screen sizes, and records what the browser saw on the way.

Screenshots at several sizes

Pick up to three sizes in one run: Desktop (1366×768), Desktop HD (1920×1080), Laptop (1280×800), Tablet (768×1024), Mobile (390×844), Small mobile (360×780), or a custom size such as 1440x900. Each size is loaded separately, in a fresh browser profile, with the matching pixel density: phones at 3×, tablets at 2×, desktops at 1×, or a density you choose.

You can capture the visible area or the full page (up to 16,000 image pixels tall), as PNG, JPEG or WebP. Under Advanced you choose when the page counts as ready (DOM ready, load event, or network idle), add an extra wait of up to 10 seconds for animations and late content, switch to a dark colour scheme, or turn the page's JavaScript off.

The preview has Fit, 100% and zoom controls, a download button and Copy image (JPEG and WebP are converted to PNG for the clipboard). With several sizes, the Compare tab shows them side by side. If you capture the same URL again later, a slider lays the new image over the earlier one.

What the browser saw

Next to the image, the inspector shows the details behind it:

  • Overview: HTTP status and redirects, when the page became ready, Largest Contentful Paint and Cumulative Layout Shift, number of requests, bytes transferred, failed requests and requests to other sites. Below that are a load timeline (first byte, first paint, first contentful paint, LCP, DOM interactive, DOMContentLoaded, load), transfer by resource type, the top hosts, and the page's title, description, canonical, language and viewport meta.
  • Resources: every request with type, status, size, duration and a waterfall bar, with filters for type, URL and failed requests only.
  • Network: final URL, protocol, the address the browser connected to, TLS version and certificate, the full redirect chain, which common security headers are present, and all response headers. Cookie values are removed; only cookie names are shown.

Every timing comes from the browser's own Navigation, Paint and LCP timing APIs. If the browser could not measure something, the value is shown as "n/a". Nothing is estimated.

A real example

A capture of https://example.com/ from the NetTools server on 6 November 2026, at Desktop and Mobile, full page, WebP:

Desktop 1366×768Mobile 390×844
Image1366×787 px, 1×, 36 KB1170×2694 px, 3×, 126 KB
HTTP status200, TLS 1.3200, TLS 1.3
First byte58.7 ms30.5 ms
Largest contentful paint148 ms88 ms
Layout shift00
Requests / transferred2 / 2.0 KB2 / 2.0 KB

example.com is about as small as a page gets, so the numbers are low. On a real site, the Resources tab is where it gets interesting: which scripts load before the first paint, which images arrive late, and how much comes from other domains.

Isolated by design

A browser that opens any address a visitor types is an obvious way into internal networks, so the capture service was built around that risk:

  • Every connection the browser makes goes through a filtering proxy: the page, redirects, images, scripts, WebSockets, even requests to localhost. The proxy resolves each name once, refuses it if any address is private, loopback, link-local (cloud metadata), carrier-grade NAT, reserved or one of NetTools' own servers, and connects only to the address it checked. DNS rebinding cannot switch the target afterwards.
  • The browser runs as a separate service that can reach nothing but that proxy, enforced by the operating system and the firewall as independent layers. Chromium's own sandbox stays on.
  • Each size gets an empty profile: no cookies, logins, cache or extensions; service workers, downloads and pop-ups are blocked.
  • Nothing you enter is executed: there is no way to supply scripts, headers, cookies or browser options. Only ports 80, 443, 8080 and 8443 are allowed.
  • Limits: one capture at a time and 12 per hour per visitor, 6 per site per 10 minutes, at most two captures running at once.

Storage and privacy

Images and results are stored under a random identifier, without your IP address or an account, and deleted after 24 hours. The Delete button removes them immediately. Copy link gives you a link you can share while the capture is kept. Your list of recent captures lives only in your browser. Exports (JSON, CSV of the resource list, copy, print) are generated in the browser from the result.

Know the limits

Each size is a single first visit from the NetTools server through the filtering proxy, so treat the timings as a lab snapshot for comparing sizes and runs, not as what your visitors experience. The browser does not scroll, click, log in or accept consent dialogs, so images that load on scroll may be missing. Some sites show challenge pages to automated browsers. Very long pages are cut at the height limit and marked as truncated.

Try it

Open the Website Screenshot & Browser Inspector and capture a page at desktop and mobile size. The guide explains every option and result, and the Redirect Checker, Page Size Checker and Web Pentest & Security Assessment go deeper on redirects, page weight and security headers.