New · Web history

The web as it was

arkibber turns the Internet Archive into answers: what a page said on any date, when it changed, and what the docs said for an older version. For people in the app, and for AI agents over MCP.

See how any website changed

Free, no account needed. Every answer links back to its Internet Archive snapshot.

Try:

Answers only history can give

“What did this pricing page say in March 2023?”

Get the page text from the closest capture, with a link to the exact Wayback snapshot.

page_at_date

“When did they change their refund policy?”

List every distinct version of a page between two dates, so you can see exactly when it changed.

page_changes

“What did the docs say for version 2.x?”

Pull the documentation as it was on a release date, for code written against an older version.

docs_at_version

One index, four ways to use it

Agent and coding-tool developers

Give your agent the web's history in one MCP call.

Four tools over MCP or REST. $29/mo with 2,000 calls included.

Builder plan →

Founders, SEOs and analysts

See how any competitor's site changed, and when.

Track 25 domains, get email digests of what changed, and run due-diligence reports.

Pro plan →

Product marketing and competitive intel

Know the moment a competitor's site changes.

Shared tracking across 150 domains with Slack and webhook alerts.

Team plan →

IP and trademark practitioners, investigators

Cited web-history timelines, ready for your file.

PDF and JSON reports with snapshot links, capture times and SHA-256 hashes.

Evidence plan →

For agents

Web history for agents

Live search and scrape tools tell your agent what the web says today. arkibber answers the questions they can't: what it said before, and when it changed. Add one MCP server and your agent gets four tools.

Works alongside Firecrawl Alexandria. Alexandria gives agents one interface to papers, developer docs, government records, official data providers and the live web. It doesn't cover how pages looked in the past. Connect both to the same agent and it can answer across the present and the past.

Get an API key Read the tutorial

arkibber is independent and is not affiliated with or endorsed by Firecrawl.

{
  "mcpServers": {
    "arkibber": {
      "type": "http",
      "url": "https://arkibber.app/api/mcp",
      "headers": { "Authorization": "Bearer ark_live_…" }
    }
  }
}

Built to support the Internet Archive

Polite by design

Cached, rate-limited fetching with an identified user agent and a contact address.

Credit where it's due

Every answer links to its Wayback snapshot, sending readers and attribution back to the Archive.

Derived data only

We serve text, changes and timelines. We don't redistribute captures in bulk.