Wayback Machine Snapshots: archive history, archived URLs and summaries

Look up the Internet Archive Wayback Machine history of any web page or domain and get it as clean JSON, CSV or Excel. The Actor uses the Archive's official public CDX API: no login, no browser, no proxies. Four modes:

Mode	You get	Typical use
Snapshots	One row per capture: date, archive link, raw-content link, HTTP status, MIME type, size and content digest	Page history, change tracking, evidence, research
Archived URLs	Every unique URL archived for a domain, host or path prefix, with its first capture	SEO (old URLs, lost pages, redirect planning), OSINT, recovering deleted content
Closest	The single capture nearest to a date you choose	"What did this page say on 1 June 2015?"
Summary	One row per input: first and last capture, years archived, months with captures per year, number of distinct versions	Domain age and history checks, due diligence, expired-domain research

Price: $0.50 per 1,000 snapshot/URL rows ($0.0005 each) and $0.003 per summary row, plus a tiny start fee. Inputs that fail are free error rows.

Features

• 🔁 Only real changes: set collapse to Only when the content changed to get one row per distinct page version (by content digest) instead of thousands of identical captures. You can also collapse to one capture per hour, day, month or year.
• 🌐 Domain-wide: match an exact URL, a path prefix (example.com/blog/*), a host, or a whole domain with all its subdomains (*.example.com).
• 📅 Date ranges (from / to), only HTTP 200 and only HTML filters.
• 🔗 Every row has a ready archive link (https://web.archive.org/web/<timestamp>/<url>) and a raw link (…id_/…) that returns the original archived bytes without the Wayback toolbar, ideal for downloading old content.
• 📄 Automatic pagination with resume keys, so large histories come back complete (up to your limits).
• 🤝 Polite by design: about one request per second with backoff, so runs are reliable and fair to the Internet Archive.

Input examples

Capture history of a page, one row per content change since 2018:

{ "urls": ["https://www.apple.com/iphone/"], "mode": "snapshots", "collapse": "changes", "from": "2018" }

All archived HTML URLs of a domain:

{ "urls": ["example.com"], "mode": "urls", "onlyHtml": true, "onlyStatus200": true, "maxResultsPerUrl": 5000 }

What a page looked like on a date:

{ "urls": ["nytimes.com"], "mode": "closest", "closestTo": "2008-11-05" }

Domain history summary for a list of domains:

{ "urls": ["apify.com", "example.com", "some-expired-domain.com"], "mode": "summary" }

Output

Snapshots / Closest rows:

Field	Description
url	The original URL as captured
timestamp, capturedAt	Wayback timestamp (YYYYMMDDhhmmss) and ISO date (UTC)
archiveUrl, rawArchiveUrl	Wayback page and raw original content
statusCode, isRedirect	HTTP status at capture time
mimeType, lengthBytes, digest	Content type, compressed size, SHA-1 content digest (same digest = same content)
urlKey	The Archive's canonical (SURT) key
requestedDate, offsetDays	Closest mode only: your date and how many days away the capture is

Archived URLs rows: url, urlKey, firstCapturedAt, firstArchiveUrl, latestArchiveUrl (always redirects to the newest capture), mimeType, statusCode.

Summary rows: archived, firstCapturedAt, firstArchiveUrl, lastCapturedAt, lastArchiveUrl, lastStatusCode, daysSinceLastCapture, yearsArchived, monthsWithCaptures, monthsWithCapturesByYear (e.g. {"2019": 12, "2020": 11}), distinctVersions (capped at 10,000; distinctVersionsCapped tells you if it was).

Every row also has input, scrapedAt and error. A SUMMARY record in the default key-value store lists the result count per input.

Example snapshot row:

{
  "input": "example.com", "url": "http://example.com/", "timestamp": "20200101000000",
  "capturedAt": "2020-01-01T00:00:00.000Z",
  "archiveUrl": "https://web.archive.org/web/20200101000000/http://example.com/",
  "rawArchiveUrl": "https://web.archive.org/web/20200101000000id_/http://example.com/",
  "statusCode": 200, "isRedirect": false, "mimeType": "text/html", "lengthBytes": 1234,
  "digest": "…", "urlKey": "com,example)/"
}

Cost examples

• Content-change history of 20 pages, about 50 versions each: 1,000 rows ≈ $0.50
• 10,000 archived URLs of a domain ≈ $5.00
• Summaries for 100 domains ≈ $0.30

Use maxResultsPerUrl (default 1,000) and maxResultsTotal to cap a run. Popular sites can have millions of captures.

FAQ

Does it download page content? No. It returns capture metadata and direct links. Use rawArchiveUrl to fetch the original content of any capture.

Why is a site missing? The site may never have been archived, or its owner asked the Internet Archive to exclude it. Excluded sites return a free error row.

Is it allowed? Yes. It uses the Internet Archive's public CDX API, made for this purpose, at a polite request rate. Please respect the Internet Archive's terms of use.

Is it affiliated with the Internet Archive? No. This is an independent tool. If you find the Wayback Machine useful, consider donating to the Internet Archive.