Wayback Machine Downloader — Recover Deleted Websites & Archive.org Data

Download Wayback Machine snapshots, recover dead websites, and scrape archive.org history — HTML, files, and metadata — in one run. This Wayback Machine downloader queries the official web.archive.org CDX API to list every archived capture of a domain, then optionally saves the raw historical HTML and files so you can restore, migrate, or analyze sites that no longer exist online.

Whether a site was shut down, a page was deleted, or you need the historical version of a competitor's store, this actor turns the Internet Archive into structured, downloadable data.

What this actor does

• 🔎 List every archived snapshot of any URL or full domain from the Wayback Machine (archive.org).
• 📄 Download historical HTML exactly as it was captured, with the Wayback banner stripped (raw id_ captures).
• 🗂️ Extract archived files — PDFs, images, and other assets — using MIME filters.
• 🕒 Filter by date range to grab a specific point in a website's history.
• 💾 Structured output — clean JSON metadata to the dataset, raw content to the Key-Value Store.

Use cases

• Dead website recovery — the site is gone, but archive.org still has it. Pull the last working snapshot and rebuild.
• Deleted page retrieval — recover a single removed URL and its historical content.
• Content migration — export an old site's pages before decommissioning a legacy platform.
• Competitor & market research — inspect how a competitor's website, pricing, or messaging looked in past years.
• Digital archiving & compliance — capture a verifiable historical record of any public web page.
• SEO & link recovery — find old URLs and content that used to rank.

Input

Field	Type	Description
targetUrl	string	Site or page to recover, e.g. example.com. Required.
mode	select	single_page, full_site (domain crawl), files_only, or snapshot_list (metadata only).
fromDate	string	Start date YYYYMMDD (optional).
toDate	string	End date YYYYMMDD (optional).
mimeFilter	array	Keep only matching MIME types, e.g. application/pdf, image/.
only200	boolean	Skip non-200 captures (redirects, errors). Default true.
limit	integer	Max snapshots to process. 0 = no limit.
saveContent	boolean	false = metadata only (cheaper). true = download HTML/files to Key-Value Store.

Example input

{
  "targetUrl": "example.com",
  "mode": "full_site",
  "fromDate": "20100101",
  "toDate": "20151231",
  "only200": true,
  "limit": 500,
  "saveContent": true
}

Output

Each snapshot is pushed to the dataset as a clean JSON record:

{
  "timestamp": "20100603215612",
  "original_url": "http://www.example.com/",
  "mimetype": "text/html",
  "statuscode": "200",
  "digest": "AB2CD3EF4GH5IJ6KL7MN8OP9QR0ST1UV",
  "captured_at": "2010-06-03T21:56:12",
  "raw_url": "https://web.archive.org/web/20100603215612id_/http://www.example.com/"
}

When saveContent is enabled, the raw historical HTML and files are stored in the run's Key-Value Store, keyed by timestamp and URL, ready to download or re-host.

How it works

The actor uses the Internet Archive's CDX Server API to enumerate captures, deduplicating identical content by digest. For downloads it requests the raw id_ capture URL, which returns the original archived bytes without the Wayback Machine's injected toolbar, JavaScript, or CSS — so the recovered HTML is clean and usable.

No login, no API key, no proxy required. Data comes straight from the public Internet Archive.

Tips

• Start with mode: snapshot_list and a small limit to preview what the archive holds before downloading.
• Large domains return thousands of captures — use fromDate/toDate and limit to stay focused and control cost.
• Use mimeFilter: ["application/pdf"] to recover only archived documents.

Pricing

Pay-per-event: you pay a small fee per run start, per snapshot listed, and per file downloaded. Run metadata-only (saveContent: false) for the cheapest previews; enable downloads only when you need the content.

---

This actor reads publicly available data from the Internet Archive (archive.org). Respect the copyright and terms of use of any recovered content.