Pricing

from $0.70 / 1,000 archived captures

Try for free

Go to Apify Store

Wayback Machine Snapshots: Archived URL History & Dead Links

Try for free

Every archived capture of a URL from the Internet Archive Wayback Machine: capture time in UTC, HTTP status, MIME type, size, content digest, replay link and raw-file link. Also finds the copy closest to a date and lists the archived URLs of a whole site.

Pricing

from $0.70 / 1,000 archived captures

Rating

0.0

(0)

Developer

Samat Makatov

Actor stats

Bookmarked

Total users

Monthly active users

4 days ago

Last modified

Categories

SEO tools

Developer tools

What to look up

mode

Optional

Three different questions, one per run. captures answers "when was this page archived and did it change" - one row per stored capture. closest answers "give me one working archived copy near this date", which is what repairing a dead link or a footnote needs: one row per URL, including a row marked found: false for URLs the archive never stored. siteUrls answers "which addresses under this site are in the archive at all" - one row per archived URL.

Type:string

Default:captures

Options:

capturesclosestsiteUrls

URLs

urls

Required

The addresses to look up, at most 100 per run. Write them as people write them: https://example.com/page, example.com/page and http://example.com/page are all accepted, because the archive indexes a canonical form of the address and the scheme rarely matters. A trailing slash does matter for siteUrls with prefix scope - https://www.irs.gov/pub/irs-pdf/ means everything under that folder. More than 100 URLs is refused with a clear message instead of a half-finished run.

Type:string[]

Date to aim at

targetDate

Optional

Used by closest. The moment you want a copy from: 2016, 2016-03, 2016-03-01 or a full 2016-03-01T12:00:00Z are all read. Leave it empty to get the newest capture the archive holds. The answer is the nearest capture in either direction, and offsetDays says how far it landed from your date - a page archived twice in ten years can answer a 2016 question with a 2011 capture, so always read that column before quoting the copy.

Type:string

How wide to search the site

matchScope

Optional

Used by siteUrls. prefix keeps the path you typed and walks down from it, host ignores the path and returns the whole host, domain adds every subdomain, which is how you find the forgotten old. and files. hosts of an organisation. A big site holds millions of archived addresses, so a wide scope needs maxItemsPerUrl and the MIME or status filters to stay useful.

Type:string

Default:prefix

Options:

prefixhostdomain

Captures from

fromDate

Optional

Earliest capture to return, as 2024, 2024-01 or 2024-01-01 (UTC). The Wayback Machine's first captures are from 1996, so leaving this empty really does mean three decades of history for an old site.

Type:string

Captures up to

toDate

Optional

Latest capture to return, same forms as above. A date alone covers the whole day. Leave both dates empty for the full history.

Type:string

Only captures of the last N hours

sinceHours

Optional

A moving window for watching a page: 24 for the last day, 168 for the last week. It replaces Captures from while it is set, so a scheduled run always asks for the fresh part of the history instead of re-reading everything. Nothing in the window is a normal answer and ends the run with a status that says so - the archive crawls a small site a few times a year, not daily.

Type:number

Minimum:1

Maximum:8760

Skip captures written by earlier runs

onlyNew

Optional

Remember which captures this run wrote in the actor's key-value store and write only unseen ones next time, keyed by URL and capture timestamp. Made for a schedule: with sinceHours it turns the actor into a watcher that reports what the archive stored since the last run. The memory belongs to the actor or task that runs, so two tasks watching two sites never shadow each other.

Type:boolean

Default:false

Keep these HTTP status codes

statusCodes

Optional

Status of the page at the moment it was archived: 200 for the ones that really hold content, 301 or 302 for redirects, 404 for pages that were already gone when the crawler came. Several values are combined with OR; leave the list empty to keep every capture. The archive stores a dash instead of a status for deduplicated revisit records, so add - when you want those too - they appear as an empty statusCode column.

Type:string[]

Keep these content types

mimeTypes

Optional

Filter by what was archived: text/html for pages, application/pdf for documents, image/jpeg, audio/mpeg, video/mp4. A family prefix also works - audio/ keeps every audio type - which is how you pull a set of lost recordings or every PDF a site once published. Empty means every type.

Type:string[]

Thin out the history

collapseBy

Optional

A busy news front page is archived dozens of times a day, so the default keeps one capture per day and a year of history fits in 365 rows instead of thousands. month or year gives a quick long-range picture, none the complete record, digest drops a capture whose content hash equals the previous one. Thinning happens inside the archive's index before the row cap is applied, so it changes which captures you see, not how many rows you pay for.

Type:string

Default:day

Options:

nonehourdaymonthyeardigest

Only captures where the content changed

changedOnly

Optional

Compare the content hash of each capture with the previous capture of the same URL and keep only the ones that differ, so a page archived 400 times with 12 real edits gives 12 rows. The first capture of a URL is always kept as the starting point and its changedFromPrevious column stays empty. A changed hash means the stored bytes differ - a new advert or a rotated banner counts as a change, so treat the column as "worth looking at", not as an editorial diff.

Type:boolean

Default:false

Newest captures first

newestFirst

Optional

The archive's index is ordered oldest first, which is what you want for "when did this page appear" questions. Switch this on and the run asks for the end of the history instead, so a small row cap shows the most recent captures - the usual choice for monitoring and for checking whether a live page is archived at all.

Type:boolean

Default:false

At most this many rows per URL

maxItemsPerUrl

Optional

Keeps one heavily archived address in a list of twenty from eating the whole run. The cap is sent to the archive with the query, so a thin URL costs one request either way.

Type:number

Minimum:1

Maximum:2000

Default:25

Maximum rows

maxItems

Optional

Hard cap on the rows this run writes, and therefore on what it costs. The run stops as soon as it is reached and says so in its status message. The archive's index hands out at most about 10000 rows for one query, so a whole-domain export is done in slices - by year with Captures from / Captures up to, or by path.

Type:number

Minimum:1

Maximum:2000

Default:50

Only these columns

fields

Optional

Keep only the named columns, in the order given - for example originalUrl, capturedAt, rawUrl. Empty = every column. queryUrl, originalUrl, found, sourceMode, url and fetchedAt are always present; name one of them here and it moves to the position you gave it. Unknown names are reported in the status message instead of being dropped in silence.

Type:string[]

Wayback Machine URL Extractor - Archived URLs

logiover/wayback-machine-url-extractor

Extract every archived URL of any domain from the Internet Archive's Wayback Machine (CDX API). Recover lost or old pages, build redirect maps and run OSINT, with date and status filters. No API key, export to CSV or JSON.

Logiover

Wayback Machine Scraper - Track Website Changes Over Time

ryanclinton/wayback-machine-search

Search the Internet Archive's Wayback Machine for historical snapshots of any website. Retrieve archived page metadata -- including timestamps, URLs, MIME types, HTTP status codes, and content hashes -- for up to 10,000 snapshots per run.

Ryan Clinton

220

Wayback Machine Scraper — Website History & Snapshots

alaudinburki/wayback-machine-snapshots

Extract every Wayback Machine (archive.org) snapshot of any URL, domain or subdomain — capture date, HTTP status, MIME type, content digest, byte size and direct snapshot links. Up to 50,000 rows, many URLs per run, $0.50 per 1,000 rows. No API key.

alaudin burki

Wayback Machine Scraper

glassventures/wayback-machine-scraper

Scrape Wayback Machine archive snapshots for any URL or domain. Get archived URLs, timestamps, status codes, MIME types. Export to JSON, CSV, Excel.

Glass Ventures

Wayback Machine Snapshots Scraper — Internet Archive History

seemuapps/wayback-machine-snapshots-scraper

List every Internet Archive snapshot of a URL, page, or whole domain. Timestamp, snapshot URL, status code, mime type, content length. No login.

Seemu Scraping

Wayback Machine Archive Scraper

andok/wayback-machine-scraper

Fetch historical snapshots of any webpage from the Internet Archive. Perfect for digital forensics and tracking deleted content.

Andok

173

Wayback Machine Scraper — Archived Snapshots

hipersoft/wayback-machine-scraper

List every Internet Archive (Wayback Machine) snapshot of a URL or whole domain: timestamp, snapshot URL, status code, MIME type and content digest. Filter by date, status and dedupe. For SEO, OSINT and historical research. No setup.

hiper soft

Wayback Snapshots — CSV, Date-Filter, Bulk JSON

knotless_cadence/wayback-machine-scraper

Wayback Machine snapshots CSV/JSON — per snapshot: timestamp, status, MIME, size, archive URL — date-filterable. CDX API, no key. 21+ runs. For competitor history-tracking + SEO recovery + brand archaeology. spinov001@gmail.com · blog.spinov.online · t.me/scraping_ai

Alex

Wayback Machine Checker (Archive.org)

bgfc97/wayback-machine-checker

Check whether URLs are archived in the Internet Archive Wayback Machine and get the closest snapshot (optionally near a target date). For OSINT, SEO, research and journalism. No key, no proxy.

Bruno

Wayback Machine Bulk Lookup

jungle_synthesizer/wayback-machine-bulk-lookup

Look up Wayback Machine snapshots for any URL or list of URLs. Returns capture timeline, optional snapshot markdown, and live-vs-snapshot diff. Date range filtering, capture limit, bulk input. Built for OSINT, journalism, SEO link-rot recovery, and legal evidence.