Wayback Machine Scraper API - Archived URLs & Website History

Search the public Internet Archive Wayback Machine CDX index for archived URLs and historical website captures. Use it for SEO recovery, redirect planning, deleted-page research, website-history timelines, OSINT, compliance research, and change investigations.

No Internet Archive API key, login, browser, or proxy is required for normal queries.

What you can search

• One URL/domain or a batch of up to 20 targets
• Exact URL, path prefix, host, or domain plus subdomains
• Capture date/timestamp range
• Archived HTTP status such as 200, 301, 404, or all statuses
• Exact MIME type such as text/html or application/pdf
• Every capture, unique content, one row per URL, or monthly/daily/hourly buckets
• Latest or earliest captures first
• Up to 10,000 total bounded output records

Empty input uses a fast deterministic exact example.org/ fixture with up to 5 successful unique-content captures. At current pricing, that first run costs at most $0.0101 ($0.0001 start + 5 × $0.002 records). A buyer-set maximum total charge reduces maxResults before the archive query. When urls contains targets, it takes precedence over the single url field. For batch input, maxResults is a total response cap split evenly across targets.

Example inputs

Latest unique versions of one page

{
  "url": "example.org/",
  "matchType": "exact",
  "statusCode": "200",
  "collapseBy": "DIGEST",
  "sortDirection": "latest",
  "maxResults": 5
}

Recover archived URLs for an entire domain

{
  "url": "example.com",
  "matchType": "domain",
  "fromDate": "2020",
  "toDate": "2025",
  "statusCode": "200",
  "mimeType": "text/html",
  "collapseBy": "URL",
  "sortDirection": "latest",
  "maxResults": 1000
}

Find historical redirects and deleted pages

{
  "urls": ["example.com/old-section", "example.com/retired-product"],
  "matchType": "prefix",
  "statusCode": "ALL",
  "collapseBy": "MONTH",
  "sortDirection": "latest",
  "maxResults": 200
}

Output

Each found capture-index row includes:

• requested target and match type;
• raw Wayback timestamp and UTC capture date;
• original archived URL;
• public replay URL and raw replay URL;
• archived HTTP status and MIME type;
• Wayback content digest and compressed record length;
• success/redirect flags;
• bounded-result disclosure;
• exact public CDX query URL, retrieval time, source link, and interpretation caveat.

No-match and source-unavailable outcomes are explicit dataset items instead of silent empty datasets. Request, HTTP-status, and response-body failures use bounded retries.

Important interpretation limits

Wayback CDX rows are archive-index evidence. They are not:

• proof that archived content is authentic, complete, current, or legally effective;
• proof of who owned or controlled a page;
• a complete record of every historical version;
• proof that a missing page was never published or captured;
• the website's current HTTP status;
• a guarantee that a replay URL still renders every original asset or script.

Captures can be delayed, missing, excluded by archive policy, replay-imperfect, or later unavailable. Content hashes show archive-record equality, not meaning or legal authenticity. Open important replay links and corroborate consequential conclusions with other evidence.

Source and operational notes

• CDX endpoint: Internet Archive Wayback Machine CDX index
• Replay service: Wayback Machine
• No-key public endpoint; response time varies with archive load and query breadth
• Broad host/domain searches can hit the selected result bound; every found row discloses when its per-target limit was filled