Wayback Machine — User Guide

This package lets your flows work with the Wayback Machine, the Internet Archive's free library of saved web pages. The Wayback Machine has been taking snapshots of websites since the 1990s, so you can look up how a page appeared on a given day, count how often it's been saved, download an old copy, or save a page right now so there's a permanent record of it.

You drag these nodes into your flow just like any others, connect a web address to the input, and read the results from the outputs. No account, login, or API key is needed — everything works out of the box.

What you can do

The nodes are grouped in the editor under Wayback Machine into Query, Retrieve, Archive, and URL sections.

A few things to know first

Web address (URL) — Wherever a node asks for a URL, put the address of the page you care about, e.g. https://example.com/about.

Timestamps (dates) — Several nodes let you target a point in time. A full timestamp is written as YYYYMMDDhhmmss — year, month, day, hour, minute, second. For example, 15 January 2020 at noon is 20200115120000. You can also give just the part you care about: 2020 means "sometime in 2020", 202001 means "January 2020". Leaving the timestamp empty almost always means "the most recent copy".

What comes back — Many nodes return a Snapshot, which bundles together:

• when it was captured,

• the original address that was saved,

• a ready-to-open link to the archived copy, and

• the status recorded at capture time (a small code where 200 means the page loaded normally).

Nodes that return several results give you a list of Captures (the same idea, one entry per saved copy) plus a count.

"No account needed," but be patient with giant sites — Very heavily-archived sites (think google.com, saved millions of times) can occasionally take a while to answer, or the archive may briefly be busy. For everyday pages, results come back quickly.

The nodes, one by one

Query — finding out what's in the archive

Check Availability

Finds the saved copy closest to a date you choose — or the most recent one if you leave the date blank. The quickest way to answer "is this page in the archive, and what's the nearest copy to X?"

• You provide: a URL, and optionally a Timestamp.

• You get back: Available (yes/no) and the matching Snapshot.

• Example: "Was bbc.com archived around New Year 2015?" → URL https://bbc.com, Timestamp 20150101.

Get Oldest Snapshot

Returns the earliest copy the archive has of a page — its first appearance in the Wayback Machine.

• You provide: a URL.

• You get back: Found (yes/no) and the earliest Snapshot.

• Example: "When was google.com first archived?" → returns a capture from 1998.

Get Latest Snapshot

Returns the most recent saved copy of a page.

• You provide: a URL.

• You get back: Found (yes/no) and the newest Snapshot.

List Captures

Returns a list of every saved copy of a page, with optional filters. Great for building timelines, tables, or "show me all the versions."

• You provide: a URL, plus optional settings:

• Match Type — how widely to search:

• exact — just this one page (default),

• prefix — every page whose address starts with what you typed,

• host — every page on that website address,

• domain — the whole domain, including subdomains.

• From / To — only include copies within a date range (leave blank for no limit).

• Limit — the most results to return (defaults to a sensible number so you never get flooded).

• Collapse — optional way to thin out near-duplicates, e.g. keep only one copy per day.

• Filter — optional condition, e.g. only copies that loaded successfully.

• You get back: a list of Captures and a Count.

• Example: "All the daily versions of a news homepage during March 2020" → Match Type exact, From 20200301, To 20200331, Collapse one per day.

Count Captures

Tells you how many times a page has been saved — without listing them all. Useful for "how well is this page archived?"

• You provide: a URL, and optionally a Match Type and From/To range.

• You get back: a Count, and a Capped flag that is "yes" if there were so many copies that the number was cut off at the limit (meaning the true total is even higher).

Get TimeMap

Returns a timeline of a page's saved copies. Similar to List Captures, presented as a straightforward chronological list of snapshots.

• You provide: a URL and an optional Limit.

• You get back: a list of Captures and a Count.

Retrieve — getting the actual content

Fetch Snapshot

Downloads the archived content of a page as it looked at a chosen time (or the latest copy if you leave the date blank). This is how you read or process the old page itself, not just information about it.

• You provide:

• a URL,

• an optional Timestamp (blank = latest copy; a date = the closest copy to it),

• a Mode:

• raw — the original page exactly as it was saved, with nothing added (default),

• replay — the page as the Wayback Machine displays it, including its viewer bar.

• You get back: the Content, a Status code, the Resolved Timestamp (the exact copy that was actually used), and the Final URL it fetched.

• Example: "Get the text of a company's homepage as it was in 2012" → URL + Timestamp 2012, Mode raw.

Archive — saving a page

Save Page Now

Asks the Internet Archive to capture a live page right now, creating a fresh permanent snapshot.

• You provide: the live URL you want archived.

• You get back: Saved (whether a snapshot was recorded), the resulting Snapshot, and a Request Status.

• Good to know: Saving is a real request to the Internet Archive and can take a moment. Because it's a free public service, it's sometimes rate-limited — if a save doesn't complete immediately, the page may still be processing, and you can look it up again a little later with Check Availability or Get Latest Snapshot.

• Example: Archive an important announcement page before it might be edited or taken down.

URL — working with archive links (no internet needed)

These two are instant helpers that don't contact the archive — they just build or read archive links.

Build Wayback URL

Turns a normal web address plus a date into a ready-to-share Wayback Machine link.

• You provide: a URL, a Timestamp, and a Mode. (Leaving the timestamp empty or * builds a "show me all saved copies" calendar link.)

• You get back: the constructed Wayback URL.

• Example: Make a link that always opens a page as it looked on a specific day.

Parse Wayback URL

The reverse: takes a Wayback Machine link and breaks it into its parts.

• You provide: a Wayback URL.

• You get back: the Original URL, the Timestamp, the Mode, and Valid (whether the link was recognized as a Wayback link).

• Example: Someone pasted an archive link and you want to pull out the original address and the date it points to.

Common recipes

"Is this page archived, and give me a link to the latest copy." → Check Availability (or Get Latest Snapshot), then feed the snapshot's link wherever you need it.

"Show me how this page looked on a specific date." → Fetch Snapshot with that date to read the content, or Build Wayback URL with that date to get a link a person can click.

"How thoroughly is this page archived?" → Count Captures for a number, or List Captures for the full breakdown.

"Archive this page now, then confirm it saved." → Save Page Now, then Check Availability to confirm and grab the fresh snapshot.

"I have an archive link — what original page and date does it point to?" → Parse Wayback URL.

Frequently asked questions

Do I need an Internet Archive account or key? No. Every node uses the Internet Archive's free public service.

Why did a lookup come back empty? Either the page genuinely hasn't been archived, or the archive was momentarily busy. Try again, or broaden your search (for example, use domain match type, or drop the date filter).

What date format should I use? Year-first: YYYYMMDDhhmmss. Partial is fine — 2019, 201906, 20190615 all work. Empty usually means "the newest copy."

Can I archive a page that's about to disappear? Yes — that's exactly what Save Page Now is for.

Is this the same as the real Wayback Machine website? Yes. These nodes talk to the same Internet Archive service you'd use at web.archive.org; they just make it available inside your flows.