Website copier: what the six real ones keep, and what they lose

A website copier saves a website to your own disk so it still renders when the original server is gone. Six are worth your time — HTTrack, wget, monolith, SingleFile, Chrome's own Save page as, and CopyAnySite — and they split into two families that fail in completely different ways.

The two families

Downloaders (HTTrack, wget -p -k, monolith) speak HTTP and nothing else. They ask the server for a file, save what comes back, rewrite the links. No layout engine, no JavaScript: whatever the page built after load does not exist as far as they are concerned. On a 2005-era static site they are perfect and free. On anything with a hero animation, a slider, lazy-loaded images or a webfont loader, they save the empty shell that JavaScript was supposed to fill.

In-browser capturers (SingleFile, CopyAnySite) read the page your browser has already rendered — the live DOM, the resolved computed styles, the fonts actually in use. Client-side rendering, pages behind your own login and geo-gated content all come along, because the browser did the hard part already.

Measured, with the original server unreachable

A copy that still fetches from the site it copied is not a copy. Every result below was re-rendered with the original origin blocked and diffed pixel by pixel against the live page — same eight pages, same renderer, three runs each:

The averages hide the interesting part. On a plain page — Hacker News, a Wikipedia article — everyone scores 99–100% and no tool deserves credit. The spread opens on pages built by a theme or a framework: on a commercial WordPress theme demo it was 98.39% against 74.64% and 65.99%, and on a heavily animated marketing page 95.99% against 65.55% and 49.03%. Reproduce any row with npm run bench:rivals in the repository.

HTTrack is not in the table because it crawls sites rather than single pages, so it cannot be scored on the same eight URLs; it belongs to the downloader family and shares wget's blind spot. If that is what you came for, read HTTrack alternatives.

Picking one

• A whole static site, hundreds of pages, unattended → HTTrack or wget -p -k. Slow, free, thorough, and blind to anything rendered.
• One article you want as a single file → monolith, or SingleFile if the page has any JavaScript in it.
• A modern page that has to look identical offline → an in-browser capturer. See how to copy a website for the step by step.
• **A copy you intend to edit afterwards** → the format matters more than the fidelity: WordPress blocks or an agent-readable kit rather than one frozen HTML file.
• Free, no card → all of the downloaders, SingleFile, and CopyAnySite's free key. Compared in copy any website free.

What no website copier can do for you

None of them can copy a page you cannot open yourself: paywalled, login-walled or IP-blocked content stays out of reach, and any web service claiming otherwise is running its own headless browser against the same wall. None of them copy the server — no PHP, no database, no API, no checkout. And none of them make a copied design yours: layout and code are copyrightable, and shipping a rebuilt page as your own is a legal problem no tool solves. Is it legal to copy a website goes through the parts that matter.

FAQ

What is a website copier?

A tool that saves a website to local storage so it renders without the original server: either an HTTP downloader that mirrors files and rewrites links, or a browser extension that captures the page after it has finished rendering.

What is the best free website copier?

For a static multi-page site, wget -p -k or HTTrack. For a modern single page, an in-browser capturer — SingleFile or CopyAnySite's free key — because both see what JavaScript rendered. Measured numbers are in the table above.

Is there a website copier online, with nothing to install?

There are URL-in, ZIP-out services, but they fetch the page from their own servers, so anything behind your login or geo-gated is invisible to them and your target URL lands in their logs. A browser extension needs neither, because the page is already open in front of you.

Does HTTrack still work?

Yes, for what it was built for: crawling and mirroring static sites. It has no JavaScript engine, so on a site whose content appears after load it saves the shell, not the page.

Can a website copier copy a site's backend?

No. Forms, logins, search, checkout and anything reading a database live on the server. A copy is the rendered front end, which is why an export you can rebuild in WordPress is usually more useful than a frozen file.

Keep reading

Clone any website — what a clone can and cannot include

Clone" promises more than any tool delivers. Here is the part of a website that genuinely clones, the part that never will, and how to tell which you need.

A Chrome extension that copies any website — how it works, and why in-browser wins

Why a browser extension copies pages that URL-based tools cannot, what CopyAnySite's MV3 extension actually does with the page, and what permissions it needs.

Copy any website's design — the layout, type and palette, not the content

How to extract a site's real design decisions — type scale, spacing rhythm, palette, grid, motion — and reuse them legitimately instead of copying the page itself.

Copy any website free — the options, and where each one breaks

Five genuinely free ways to copy any website, what each keeps, where each falls apart, and the measured fidelity of every one with the original server offline.