One file with everything inside it: text, images and styling. It opens in any browser, offline, long after the original has changed or disappeared.
A real browser loads the page, so modern layouts, web fonts and images come out right.
Everything is inside one file. Move it, email it or file it anywhere, and it still opens.
Known cookie banners are accepted first, so you save the page, not the overlay. The record says so.
The address, and every redirect after it, is checked before anything is fetched.
A real Chrome browser loads the page and accepts a known cookie banner.
It waits for the page to finish loading, including content further down.
Images, styles and fonts go inside one file. Scripts are removed.
The preview is taken from the saved file, so a broken capture looks broken straight away.
Text stays text, so you can search it, select it and quote it. Links keep their addresses. Images, fonts and styles are inside the file, so it does not depend on the original server being there.
Scripts are removed and the file carries a policy that stops it reaching the network, so an archive opened years later cannot phone home. A short note at the top says where it came from.
<!-- Archived by Archivepanel (https://archivepanel.com) requested url: https://example.org/terms page opened: 2026-03-02T09:12:02Z page saved: 2026-03-02T09:12:31Z viewport: 1440x900 browser: Chromium 154.0.8037.57 saved with: single-file-cli 2.16.0 -->
The saved page is the archive. Alongside it you get a preview to recognise it by, a plain-text version to quote and compare, a WARC for archives that accept nothing else, and a manifest that lists a SHA-256 for every file.
Public pages only, one at a time. Pages behind a login are not captured.
Save the supplier's refund terms as they read today, before anyone has a reason to edit them.
Keep the version you read, so a footnote still points at something after the article is moved or rewritten.
A dated copy of the advertisement or the pricing page, in a form you can hand to someone else.
Five files, all counted towards your storage:
.html file. This is the archive.If one of the last four cannot be made, the capture is still kept and the record says which is missing.
So the archive cannot change after it is saved, and cannot contact anything when it is opened. The saved file also carries its own rule against network access, wherever it ends up. Video and audio are not captured.
Where a known consent platform is recognised, its own accept button is pressed before the page is saved. That also reveals pages that are hidden until you consent. There is no guessing: anything that is not a recognised banner is left alone, so no part of the real page is removed by mistake. The record states that consent was accepted automatically.
Problems another try could fix, such as a timeout, are retried automatically after 2 minutes and again after 10. You get one alert if the last attempt fails too. Problems no retry can fix, such as a page over the size limit, are reported at once. A capture you stop writes nothing and costs no storage.
No. Nothing rewrites a snapshot. Capturing the address again files a new snapshot beside the old one, and deleting one takes your password.
Each capture has 150 seconds, the saved page can be up to 48 MB, and pages are loaded in a 1440 × 900 window. Every limit is listed in the documentation.
Start free with 100 MB of storage. No card required.