Updated: 2026-10-27
You want the whole archived site on your own disk, not one page at a time from the browser. And you would rather run it yourself than pay anyone. Fair enough. That is what this tool is for.
The Wayback Press downloader is free and open source. It pulls an archived site from the Internet Archive’s Wayback Machine and saves it as HTML files or as a WordPress import file. No AI, no account with us, no charge. You run it on your own machine.
The open-source repository is being prepared; this page will link to it once it is public.
What you get
- The full archived site, not a preview: every page the archive index lists for the domain that the tool can fetch.
- Two output formats. Plain HTML files you can open or host as they are, or a WordPress import file that the standard WordPress Importer can read. That importer brings in posts, pages, comments, custom fields, categories and tags (WordPress Importer, wordpress.org).
- The source code. Read it, change it, run it in your own scripts.
Requirements, install steps and command options are documented in the tool’s README, published with the repository when it is public.
Look before you download: the Wayback CDX API
You can check what the archive holds for a domain before downloading anything. The Internet Archive’s CDX server answers a plain web request with one row per capture: the URL key, timestamp, original URL, content type, status code, digest and length. Only url= is required, output=json returns JSON, and fl= picks the fields (Wayback CDX Server API, Internet Archive on GitHub). For example:
https://web.archive.org/cdx/search/cdx?url=example.com&output=json&fl=timestamp,original,statuscodeThe same README says each query has a maximum result count and recommends its pagination API for large queries. Two of its filters save work before you download anything: filter=statuscode:200 keeps only the captures that actually returned a page, and collapse=digest folds away repeats of the same unchanged page so you are not fetching ten identical copies. Trim the index first, then pull.
The flow from there:
the archive’s saved pages → HTML files or a WordPress import file
If you are new to the archive, start with the basics: What is the Wayback Machine? Then come back.
Why raw downloads are unstable (and why that is not a bug)
The tool copies what the archive holds, faithfully. The trouble is what the archive holds:
- Mixed snapshots. Different pages captured years apart, so the menu, header and content may not match.
- Missing assets. Stylesheets and images that were never archived. The Internet Archive’s own help pages say not every archived date is complete (Internet Archive Help Center).
- Junk and spam pages picked up over the years.
- Duplicate versions of one page under slightly different URLs.
- Wrong-era content if the domain changed owners and topics.
A downloader with no judgement cannot tell which of those you wanted. So you get all of it, and the cleaning is yours. For a small site, that is an evening’s work and the free tool is the right choice.
Free vs paid, honestly
| Free downloader | Paid AI-checked restore | |
|---|---|---|
| Runs on | Your machine | Our pipeline |
| Output | HTML or WordPress import file, as archived | Clean HTML or an editable WordPress site |
| Snapshot per page | Picked automatically | Best one chosen per page, checked by an LLM |
| Layout breaks | Left as found | Fixed or flagged |
| Junk, spam, thin pages | Included | Dropped and listed |
| URL structure | As archived | Kept, with a 301 map for anything that moves |
| Human review | You | A human reviewer before delivery |
| Price | Free | From $49 per site, up to 300 URLs (pricing) |
The free tool works. The paid difference is the LLM quality check and the done-for-you part, nothing hidden behind a paywall. If you want to see what a domain has before choosing either, scan a domain in the Wayback Machine with the free online scan: old design, URL structure and a few sample pages.
When the clean-up is the part you don’t want to do, the AI-checked restore is the paid route. Prices depend on the size of the site; compare plans.
Other downloaders exist too, free and paid, online and command-line. We compare them in our roundup of sites like the Wayback Machine.
Fine print
- Content rights. Downloading is not permission to republish. If you restore a site, you need the rights to its content; our terms say the same for the paid service.
- Be polite to the archive. It is a public, non-profit resource. Large pulls should go slowly; the CDX documentation gives no request-rate figure, only a cap on results per query (Internet Archive, CDX server documentation).
- Not affiliated. Wayback Press is not affiliated with the Internet Archive, and not affiliated with Shift8 Web or its open-source tool that shares the “Wayback Press” name.
- No reviews yet. The paid service is new; first deliveries become case studies only with the client’s written consent.
The downloader gives you every file the archive kept. Deciding which ones were your site is still the interesting part.