Every Wayback Machine capture of a URL or a whole domain, as JSON
The Wayback Machine API returns the Internet Archive's captures of web pages as JSON: every archived capture of a URL with its timestamp, HTTP status, MIME type, content digest, size and a ready web.archive.org snapshot link, the closest capture to any date, a page's first-seen and last-seen dates with captures per year, every archived URL of a whole domain, and an archived-or-not check for up to 20 URLs in one call.
- endpoints
- 5
- credit per call
- 1
1 credit per call≈ $1/1,000 calls1,000 free credits, no card
5 active endpoints. Every call is 1 credit.
- POST/wayback-machine/v1/snapshots1 credit
- POST/wayback-machine/v1/available1 credit
- POST/wayback-machine/v1/history1 credit
- POST/wayback-machine/v1/domain_captures1 credit
- POST/wayback-machine/v1/batch1 credit
Run the call yourself, on the real endpoint
Change a parameter and press Try it. The response on the right is what your code will receive.
Wayback Machine API
4 of 5 endpoints, ready to run
Every archived capture of a URL, with status, digest and snapshot link
// Press "Try it" and this pane shows exactly what the // live site returned this second — including an empty // result, if that is the truth. No key, no account.
What Wayback Machine endpoints does ReefAPI ship?
5 live read endpoints. Read-only data API: no writes, no account actions, no dashboard access on the target site.
Every parameter, every allowed value →
Missing a Wayback Machine endpoint, or need a source we don't have yet?Contact us real people · same-day reply.How the Wayback Machine API works
Wayback Machine is a normal ReefAPI surface — the same four rules that hold for every other engine on the key.
No OAuth app, no request signing, no per-site account. One key covers all 746 engines.
Every route is a POST with a JSON body. Parameters are validated against the published schema before anything is charged.
Credits, not seats. Failed and blocked calls are never charged, and nothing is cached: every call reads the source live.
One envelope everywhere. meta carries latency_ms, record_count and the endpoint that answered.
What did this page say on a date, and when did it change?
Check that a page was archived near the date you care about, read its lifespan, then list the captures at which its content actually changed.
{"urls": ["github.com", "python.org", "no-such-host-reef-probe.invalid"], "timestamp": "20150101"}Up to 20 URLs in one call. On 2026-11-21: 2 of 3 archived, github.com closest to 2015-01-01 at 2015-01-02 01:57:09 UTC, python.org at 00:17:31 UTC, the unknown host archived false; 5.4 s, 1 credit.
{"url": "hirecream.com/blog/careerbuilder-review-faqs"}first_seen and last_seen read exactly from the index, span in days, captures per year, status codes and MIME types; 5 to 7 s on 2026-11-21.
{"url": "<the url>", "collapse": "digest"}One row per content change, each with timestamp, HTTP status, MIME type, SHA-1 digest, size and a ready snapshot_url; page on with meta.resume_key. Old calls to POST /web-archive/v1/... keep working unchanged.
A dated list of the archived versions of a page, with a link to each copy on web.archive.org and the dates on which its content changed.
curl -X POST https://api.reefapi.com/wayback-machine/v1/snapshots \
-H "x-api-key: $REEF_KEY" \
-H "content-type: application/json" \
-d '{"url":"github.com","limit":10,"collapse":"digest"}'{
"ok": true,
"data": { … },
"meta": {
"api": "wayback-machine",
"endpoint": "snapshots",
"mode": "live",
"latency_ms": …,
"record_count": …
},
"error": null
}What a Wayback Machine field actually contains
Fields that read differently from how they look. Each row was measured on live Wayback Machine data on 2026-11-21.
| Field | What it holds |
|---|---|
| timestamp | 14 digits, YYYYMMDDhhmmss, in UTC ("20150801011457"). from, to and timestamp accept an ISO date or any 4 to 14 digit prefix of it. |
| datetime | The same moment as ISO 8601 with Z ("2015-08-01T01:14:57Z"). |
| snapshot_url | The archived copy: https://web.archive.org/web/<timestamp>/<original>, ready to open. |
| statuscode | The HTTP status the crawler got at capture time, as text: "200", "308", "302". Revisit records (mimetype warc/revisit) carry "-": the crawler saw the same content as before. |
| digest | SHA-1 of the captured content in base32. The same digest means the same content; collapse=digest keeps one row per change. |
| length | Size of the stored record in bytes (5398, or 428 for a revisit record), not the page weight in a browser. |
| available / closest | available false with closest null means no capture was found; this is a normal answer, not an error. closest is the nearest capture in either direction, so it can be after the date you asked for. |
| first_seen, last_seen (history) | Read exactly from the index. years, status_codes and mime_types count one capture per day, capped at 1,000 days (sample_capped says when the cap was reached). |
| has_more, resume_key | snapshots and domain_captures page by an opaque token: pass meta.resume_key back as resume_key for the next page. There is no page number or total. |
Captures are what the Internet Archive's crawlers reached. A page with no capture may simply never have been crawled.
Wayback Machine coverage, measured
Every number below comes from live Wayback Machine calls on 2026-11-21.
aprendehebreo.com: 100 captures on the first page (from 2015-08-01), has_more true with a resume_key, 1.9 s. lastfollowed.com: all 9 captures, including a 308 redirect capture, 1.5 s.
timestamp (14 digits, UTC), datetime (ISO), original URL, urlkey, mimetype, statuscode as text ("200", "308"; "-" on revisit records), SHA-1 digest, length in bytes and snapshot_url (https://web.archive.org/web/<timestamp>/<url>).
example.com: closest capture 2026-11-21 11:49:08 UTC, the same day, 3.5 s. A host that was never archived: available false, closest null, 4.0 s, not an error.
hirecream.com: 357 archived URLs across the domain and subdomains, one row per URL with first capture, status and snapshot link, from a 2013-07-16 homepage capture; 12.6 s and 27.8 s on two runs.
batch: up to 20 URLs per call, archived true or false per URL with the closest capture; 3 URLs answered in 5.4 s for 1 credit.
The Wayback Machine only holds pages its crawlers reached, so no capture is not proof a page never existed. A filter must be written field:regex (e.g. original:.*blog.*); a bare word was refused with HTTP 400. Calls that read a whole domain take 10 to 30 seconds. Common Crawl's monthly crawl index, with the WARC location of each capture, has its own page: /common-crawl-api.
What people build with Wayback Machine
The jobs this data is most often used for.
endpoints
credit per call
Prove what a page, price or terms of service said on a given date, with a link to the archived copy.
Find when a website first appeared and when it was last archived, for due diligence on a domain.
List every page a website ever had, including pages that are gone today, for SEO recovery and redirects.
Check whether a list of URLs was archived before a cut-off date, 20 at a time.
What Wayback Machine data costs
The cheapest call here is 1 credit, so $15/mo (Pro) buys 10,000 of them — $1.50 per 1,000 credits. Credits roll over and never expire, and failed or blocked calls are not charged.
- 1,000 free credits on signup, no card
- One key, all 746 APIs, one credit pool
- Failed and blocked calls are never charged
- Credits roll over and never expire
Call it in two lines
Sign up, get 1,000 credits and one key that works on every engine. Then this is the whole protocol.
curl -X POST https://api.reefapi.com/wayback-machine/v1/snapshots \
-H "x-api-key: $REEF_KEY" \
-H "content-type: application/json" \
-d '{"url":"github.com","limit":10,"collapse":"digest"}'import os
import requests
REEF_KEY = os.environ["REEF_KEY"]
r = requests.post(
"https://api.reefapi.com/wayback-machine/v1/snapshots",
headers={"x-api-key": REEF_KEY},
json={
"url": "github.com",
"limit": 10,
"collapse": "digest"
},
timeout=90,
)
print(r.json()["data"])Have a question? We got answers.
The questions people actually ask before wiring up Wayback Machine.
Is there a Wayback Machine API?▾
Yes. The wayback-machine API on ReefAPI returns the Internet Archive's Wayback Machine captures as JSON, no archive.org account needed: every capture of a URL, the closest copy to a date, first and last seen, every archived URL of a domain, and an archived-or-not check for up to 20 URLs per call.
How do I get the archived version of a page on a specific date?▾
Call available with the URL and timestamp (YYYYMMDD or YYYY-MM-DD). It returns the closest capture and its snapshot_url. On 2026-11-21 github.com asked for 2015-01-01 returned the capture of 2015-01-02 01:57:09 UTC. Closest can fall after your date; to stay on or before it, call snapshots with to set to that date and take the last row.
How do I find the dates a page actually changed?▾
Call snapshots with collapse=digest. It drops consecutive captures with identical content, so each row is a new version of the page. timestamp:4, :6 and :8 give one capture per year, month or day instead.
Can I list every page of a website that was ever archived?▾
Yes, with domain_captures. It returns one row per archived URL across the domain and its subdomains, with the first capture, HTTP status and snapshot link: 357 URLs for hirecream.com on 2026-11-21, back to a 2013 homepage capture. Pass status 200 to keep only pages that loaded, or "!200" to see the failures.
Can I check many URLs at once?▾
Yes. batch takes up to 20 URLs as a JSON array (or a comma-separated text of plain URLs) plus an optional timestamp, and returns archived true or false with the closest capture for each. 3 URLs answered in 5.4 seconds on 2026-11-21. A JSON array written inside a text string is split at its commas with the quotes kept, and those URLs then come back archived false; send a real array.
How fast is it?▾
Measured on 2026-11-21: snapshots 1.5 to 4.7 seconds, available 3.5 to 4.0 seconds, history 5 to 7 seconds, batch of 3 URLs 5.4 seconds, a whole-domain listing 12.6 and 27.8 seconds on two runs.
How is this different from the Internet Archive API?▾
This API reads the Wayback Machine, the archive of captured websites. ReefAPI's internet-archive API reads archive.org's catalogue of items: books, audio, films and software, with their files. Use this one for "what did this web page say in 2015".
Do old web-archive calls still work?▾
Yes. POST /web-archive/v1/snapshots, /available, /history, /domain_captures and /batch keep working unchanged. The call for the Wayback Machine alone is POST /wayback-machine/v1/<action>, with the same parameters.
What is the Wayback Machine API?▾
Wayback Machine API is a ReefAPI endpoint group for every internet archive capture of a url or domain: timestamps, status codes, snapshot links, first and last seen. It returns live JSON through POST requests under /wayback-machine/v1.
Is the Wayback Machine API free to try?▾
Yes. ReefAPI starts with 1,000 free credits, no card required. Wayback Machine calls use the same shared credit balance as every other ReefAPI engine.
Do I need a Wayback Machine login or account?▾
No login to Wayback Machine is needed for the API response. You call ReefAPI with your x-api-key header, and the playground can run live examples before you create a production key.
How fresh is the Wayback Machine data?▾
The page example is captured from a live snapshots call, and production requests fetch live data through ReefAPI rather than a static sample.
How many credits does the Wayback Machine API use?▾
Wayback Machine actions currently cost 1 credit per successful call. Failed or blocked calls are free. All APIs draw from one credit pool.
Can I call Wayback Machine from an AI assistant or MCP client?▾
Yes. Connect ReefAPI once through MCP and your assistant can call wayback-machine actions with the same key, credit pool and JSON envelope used by normal REST requests.
15 Developer Tools APIs on the same key
One key, one credit pool, one response envelope. If you are pulling Wayback Machine, you are one call away from the rest of the category — no second contract, no second integration.
Need something this API does not do?
Name the endpoint, the field, or a source we do not carry yet. We ship new APIs every week and you would be first to get the key. Real people read every message and reply the same day.
Try it on your own data before you pay anything
The call above is the real endpoint, not a recording. A free key gives you 1,000 credits, the other 745 APIs, and the same envelope everywhere.
Endpoints, parameters and credit costs on this page are read from the live catalog and cannot drift from what the API accepts. Field notes were captured on 2026-11-21.