From a pasted list to a zip, step by step.

1. Your list gets cleaned up first

Paste anything. Bare domains, full URLs with query strings, a column copied out of a spreadsheet with the quotes still attached, a markdown list, a CSV export where the URL is the third field. Every line is split on commas and tabs, stripped of stray punctuation, and given a scheme if it is missing one.

Duplicates collapse to a single render so you are not charged twice for the same address. Anything that is not a public HTTP or HTTPS address is dropped before the batch is created, and the count of what was dropped comes back with the response.

2. Each page opens in a real browser

Rendering happens in Chromium, not in a fetch and a HTML parser. JavaScript executes, web fonts load, images decode, and single page apps finish drawing before the shutter goes.

Before capture the page is scrolled from top to bottom once, so lazy loaded images and anything gated behind an intersection observer is actually in place. Animations and CSS transitions are paused so a carousel mid slide does not end up as the shot. Consent walls from the common platforms are hidden, along with the generic hand rolled banners that follow the same patterns.

3. The engine is built to not get walled

A stock headless browser gets a bot challenge on a meaningful share of real sites, and a bot challenge is not the screenshot anyone asked for. The renderer runs on Fortress, a Chromium build that corrects its fingerprint in the engine rather than by patching JavaScript at runtime, which is what the detectors look for.

You do not configure any of this. It is the default, and there is a stock Chromium fallback if the stealth engine is unavailable on a given host.

4. Results are packaged as you go

Renders run in parallel and the results page fills in as each one lands, so a five hundred URL batch is useful long before it is finished. When the batch settles you get a zip with one image per URL, named after the domain and its position in your list.

Alongside the images sits a manifest.json with the page title, the URL after redirects, the HTTP status, the rendered page size in CSS pixels and the byte size of each file. If anything failed there is a not-rendered.csv listing each failure and its reason.

5. Nothing sticks around

Images and the packaged download are deleted 14 days after the batch was created. There is no archive to browse later, which is deliberate: this service renders pages, it does not keep a copy of the web.