Wayback Machine Scraper — Archive.org Snapshots & URL History

Every Wayback Machine (archive.org) capture of a URL, folder, host or domain: timestamp, status, MIME type, digest and archive links. Unique versions, daily/yearly sampling, unique URLs, per-site summaries, optional archived page Markdown. $2 per 1,000 snapshots.

See how any website looked in the past — every archived capture from the Internet Archive's Wayback Machine (archive.org) as clean rows. Paste URLs or domains and get the full historical snapshots list from the official Wayback CDX API: capture time, original URL, HTTP status, MIME type, size, content digest and ready-to-open web archive links (with and without the Wayback toolbar). Look up one page, a whole folder (example.com/blog/*), a host, or an entire domain with all subdomains. Cut the noise with unique content versions (default — only captures where the page actually changed, flagged with changedFromPrevious), one capture per hour, day, month or year, or unique URLs to list every page ever archived on a site — great for recovering lost content, finding old URLs for redirects, and SEO expired domain research. Filter by date range, HTTP status (200, 2xx, !404) and MIME type (text/html, application/pdf, image/*). Summary mode gives one row per site with first and last capture, total captures, captures per year, missing years and longest gap — the quick domain history check before you buy a domain or vet a company. Optionally pull the archived page content as LLM-ready Markdown and text. Built for compliance and legal evidence, competitor price and copy tracking, content recovery, link building and AI agents. Polite by design: paced requests, automatic retries on archive.org rate limits and timeouts. $2 per 1,000 snapshots (the most-used Wayback actor charges $3.50), $0.004 per page summary ($0.01 for a whole-domain summary), with no compute bill on top. URLs that fail or have no captures are never charged.

What data does it return?

Every run outputs clean, schema-validated rows like these:

inputUrl	capturedAt	statusCode	mimeType	changedFromPrevious	archiveUrl	contentTitle
https://apify.com/store	2019-07-04T11:02:25Z	200	text/html		https://web.archive.org/web/20190704110225/https://apify.com/store	Store · Apify
example.com	2015-07-17T01:53:29Z	200	text/html	true	https://web.archive.org/web/20150717015329/http://www.example.com/	Example Domain
*.apify.com	2023-06-10T23:36:51Z	301	text/html		https://web.archive.org/web/20230610233651/https://www.apify.com/.well-known/ai-plugin.json

How much does it cost?

Transparent pay-per-event pricing — you only pay for what you get, and the actor always respects your run budget: set a maximum total charge and it stops cleanly at the cap instead of overrunning.

Event	Price	Charged	Worked example
snapshot	$0.002	per archived capture row delivered	Unique versions of one page over 10 years (~300 rows) = $0.60; 1,000 URLs of a domain = $2.00
url-summary	$0.004	per summary row for an exact URL (summary mode)	Archive history of 500 home pages = $2.00
domain-summary	$0.01	per summary row for a prefix, host or whole domain (scans the archive index)	Capture history of 100 entire domains = $1.00
page-content	$0.004	per archived page fetched as Markdown/text (only when Fetch archived page content is on)	10 snapshots with content = $0.02 + $0.04
apify-actor-start	$0.00005	per run per GB of memory (platform start fee)	One run at up to 1 GB = $0.00005 (charged as one event)

Input options

Field	Type	Required	Default	Description
urls	array	yes	—	Pages, folders or domains to look up in the Wayback Machine, e.g. example.com, https://example.com/pricing. Shortcuts: example.com/blog/* = everything under that path, *.example.com = the whole domain with subdomains.
matchType	string	no	"exact"	Exact: that one URL. Prefix: every URL starting with it (a folder). Host: every URL on that host name. Domain: the host and all its subdomains. Prefix/host/domain queries are slow on archive.org's side (20–40 s per index page).
outputMode	string	no	"snapshots"	Snapshots: one row per archived capture. Summary: one row per input — first and last capture, total captures, captures per year, missing years (charged per summary, not per capture).
collapse	string	no	"digest"	Busy pages are captured thousands of times with identical content. Unique versions (default) keeps a capture only when the content changed. Hourly/daily/monthly/yearly keep the first capture of each period. Unique URLs keeps one row per distinct URL (best with prefix/host/domain to list every archived page of a site). None returns every capture.
from	string	no	—	Only captures on or after this date: 2015, 201505, 20150501 or 2015-05-01.
to	string	no	—	Only captures on or before this date (the whole period is included: 2018 means up to 31 Dec 2018).
statusFilter	string	no	—	Only captures with these HTTP status codes, comma-separated: 200, 2xx, 200,301 — or exclude with !404, !5xx. Empty = all (includes redirects and errors).
mimeFilter	string	no	—	Only captures of these content types: text/html, application/pdf, image/* — or exclude with !image/*. Empty = all.
newestFirst	boolean	no	false	Exact URLs only: return the most recent captures (up to the max per URL) newest first, instead of the oldest first.
maxSnapshotsPerUrl	integer	no	1000	Stop after this many snapshot rows for each input. Each row is charged, so this is also your cost cap per URL.
fetchContent	boolean	no	false	Download the archived HTML of snapshots and add the page title and main content as Markdown/text (charged per page, see pricing). Only HTML captures with status 200 are fetched.
maxContentPages	integer	no	10	Fetch content for at most this many snapshots in the run (the first ones delivered). Failed fetches don't count.
contentFormat	string	no	"markdown"	What to include when fetching content.
maxIndexPagesPerUrl	integer	no	10	Prefix, host and domain queries walk archive.org's index page by page (each ~3,000–7,000 captures, 20–40 s). Stop after this many pages per input, even if fewer snapshots than the max were found (e.g. with tight filters on a huge site).
proxyConfiguration	object	no	{"useApifyProxy":false}	Not needed normally — archive.org's CDX API is public. Turn on Apify Proxy only if your runs keep hitting rate limits.

Limits & honest notes

• Data comes from the Internet Archive's public Wayback CDX API. Captures the archive doesn't have (never crawled, excluded by the site owner, or very recent) can't be returned; such inputs get a free row saying so.
• archive.org is a non-profit and sometimes slow: exact-URL lookups take about a second, but prefix/host/domain lookups take 20–40 s per index page (each page covers several thousand captures). The actor paces its requests (about one per second, two URLs in parallel) and retries rate limits (HTTP 429), server errors and timeouts with backoff; a URL that still fails gets a free error row.
• Prefix/host/domain results come in URL order (then time), not purely by date. Newest-first is available for exact URLs only. These queries stop after 10 index pages per input by default (raise Max index pages to go deeper). On multi-URL results, collapsing works per URL: "one per day" keeps one capture per day for each URL.
• Unique-versions mode compares each capture with the previous one (archive.org's digest collapse), so a page that changes back to an earlier version shows up again — that is a real change.
• Summary counts for exact URLs come from the Wayback Machine's capture calendar (all captures, monthly precision for date ranges). With status/MIME filters, or for prefix/host/domain, they are counted from the index (up to 100,000 captures or the index page cap; scanComplete says whether the count is complete).
• Archived page content is extracted from the raw capture as archived; pages that were built by JavaScript may come out nearly empty, and images/CSS are not downloaded. Content is fetched only for HTML captures with HTTP 200 and is capped at 200,000 characters.
• Respects your run budget: set a maximum total charge and the run stops cleanly at the cap, with no row charged twice.

Related ActorForge actors

• Website Change Monitor — track changes from now on
• Domain WHOIS & DNS Lookup
• Website to Markdown
• SEO Audit Crawler

Changelog

• 2026-10-19 — Initial release: Wayback CDX snapshots for exact/prefix/host/domain, unique versions and time/URL collapsing, status/MIME/date filters, change flags, per-site summary mode, optional archived page content as Markdown.

---

Built and maintained by ActorForge. Reliability first: every actor in the portfolio is monitored by an automated nightly self-test, and breakages are typically fixed the same day.