Resume
← Blog

2026.11.05 · 4 min · browsers · tooling · open source

Save a website before it disappears.

ArchiveWeb.page is a free, open-source browser extension that records the pages you open and lets you download them as one WACZ file that plays back like the live site. How it captures, what it misses, and where the archive actually lives until you download it. As of 2026.11.

Sites vanish without warning, and nobody else may have saved the one you care about. A recent post, If you love a website, back it up, makes the case after a fansite from 2004 died and its domain started serving gambling ads. It calls one tool "by far the easiest": ArchiveWeb.page, a browser extension that records what you browse and hands it back as a single file that replays like the live site.

Watch the 60-second version on Instagram.

It isn't new. Webrecorder announced it on 2021.01.18, and the latest release is v0.17.1 (2026.09.04 on GitHub, 2026.09.05 on the Chrome Web Store). It's AGPL-3.0, and no product page lists a price; "free" rests on Webrecorder's general open-source commitment and the store listing, which is well supported but not stated outright.

It records as you browse

ArchiveWeb.page records a tab's network traffic, so it saves what actually loaded, not a re-download of the URL. That's the difference from Save As: a page that fetches its content as you scroll or click gets captured in the state you saw it.

OPENED · SAVEDNOT OPENEDHOMEFLEXBOXGRIDSELECTORSCOLORSBROWSER STORAGEINDEXEDDB · 3 PAGESINVENTED SITE · IT RECORDS, IT DOESN'T CRAWLFIG. 01 / OPENED PAGES ARE SAVED

The other side of that is the honest limit. It doesn't crawl. It saves every page you open, and the parts of those pages you click open. Pages you never visited never loaded, so they aren't in the archive. Webrecorder points whole-site crawling to a different product, Browsertrix. "An entire website", in the post's words, means visiting each page.

The steps today:

  1. Install it from the Chrome Web Store.
  2. Open the page, click the extension icon, press START. Optionally name a new archive under "Record To".
  3. Browse, and click everything you want kept. The banner means it's recording.
  4. Click the icon, press STOP.
  5. Open the archive, choose Download, then "Download All as WACZ".
  6. On replayweb.page: Choose File, Load.

The button labels come from Webrecorder's undated guide; one third-party guide says "Save To" and "Start Archiving" instead, so check them on v0.17.1.

One file, and it isn't uploaded

Until you download, the archive is not a file. It lives in the browser's storage (IndexedDB). It can be deleted or downloaded there, there's no usage tracking, and it stays private unless you opt in to sharing. Whether uninstalling the extension deletes it isn't documented; assume it does, and download.

IN BROWSERINDEXEDDBDOWNLOADsite.waczWARCWARCINDEXTEXT INDEXREPLAYREPLAYWEB.PAGECSS Notesdisplay: flex;ON YOUR DISK · NOT UPLOADEDINVENTED SITE · FILE NAME IS AN EXAMPLEFIG. 02 / FROM BROWSER TO FILE

The download is WACZ or WARC. WACZ is a zip that holds WARC files plus indexes, which lets the player load it on demand, so Webrecorder recommends it for replay:

site.wacz
site.wacz   a zip of
  ├─ WARCs
  ├─ index
  └─ full-text index

Open it on ReplayWeb.page and the site is back. The file is "not uploaded anywhere", in ReplayWeb.page's docs.

What people get wrong

  • "The Wayback Machine has it." Maybe. The post's whole point is not to wait for someone else to have saved the page you care about.
  • "Save As is enough." A static save misses what a page loads as you use it. Recording network traffic catches that. (This comparison is my reading, not a Webrecorder claim.)
  • "It saves the whole site." It saves what you opened. Click the tabs, menus and accordions you want kept, and wait for the yellow status to clear before moving on. Autopilot scrolls social and infinite-scroll pages for you, with no guarantee.
  • "It replays with no internet." The installed ReplayWeb.page app (PWA) and the desktop app replay offline. The replayweb.page website itself isn't confirmed to.
  • "It works anywhere." It needs Chrome 87+, Edge 87+ or Brave. Firefox and Safari aren't listed. A standalone desktop app exists, and the troubleshooting guide suggests it when a PDF viewer extension blocks capture.

One caution of my own: Facebook's capture behaviour supports only the logged-in UI, so recording runs in your session. No doc says archives contain private data, but treat a logged-in archive as private before you share it.

On a client project I'd record the third-party docs and dashboards we depend on before a migration, and keep the WACZ next to the decision record that cites them.

The short version

  1. ArchiveWeb.page is a free, AGPL-3.0 browser extension (Chrome, Edge, Brave) that records a tab's network traffic as you browse.
  2. It saves every page you open and the parts you click open. It doesn't crawl; that's Browsertrix.
  3. Until you download, the archive lives in browser storage, not in a file.
  4. Download it as one WACZ: a zip of WARCs plus indexes.
  5. ReplayWeb.page opens the file without uploading it; its installed app replays offline.

Sources: If you love a website, back it up · ArchiveWeb.page, install requirements · GitHub, webrecorder/archiveweb.page · Introducing ArchiveWeb.page · GitHub releases · Chrome Web Store · Webrecorder, ArchiveWeb.page · Webrecorder, developer tools · Privacy policy · Download entire collection · Announcing WACZ 1.0 · ReplayWeb.page user guide · Loading archived items · Create collection · Initiate session · Capture session · Terminate session · SUCHO plugin instructions · Browsertrix · Autopilot · Behaviors · Offline use · Common errors · Hacker News