Wayback

Wayback Machine

Query the Internet Archive's Wayback Machine for archived web pages.

When to Use

• User wants to see an old/archived version of a web page
• Checking if a URL/domain was previously active
• Retrieving historical page content
• Searching for all snapshots of a URL
• Comparing page content over time

Requirements

No API key required — Wayback Machine is free and public.

Commands

search — Find snapshots

./scripts/wayback_search search "https://example.com" [-n 25] [--from 20200101] [--to 20231231]

check — Quick availability check

./scripts/wayback_search check "https://example.com"

latest — Fetch latest snapshot content

./scripts/wayback_search latest "https://example.com" [-f json]

get — Fetch snapshot at specific timestamp

./scripts/wayback_search get "https://example.com" -t 20200101120000

Options

| Option | Description | |--------|-------------| | -n, --limit | Max results for search (default: 25) | | --from | Start date YYYYMMDD | | --to | End date YYYYMMDD | | -s, --status | Filter by HTTP status (e.g. 200) | | --collapse | Dedupe by field (e.g. digest, timestamp:6) | | -m, --match | URL match: exact, prefix, host, domain | | -t, --timestamp | Snapshot timestamp for get/check | | --raw | Return raw HTML instead of extracted text | | -c, --max-chars | Max content chars (default: 10000) | | -f, --format | Output: json (default) or table |

Output Format

search — list of snapshots:

[
  {"timestamp": "20200101120000", "original": "https://example.com/", "statuscode": "200", "mimetype": "text/html", "digest": "ABC123"}
]

latest/get — extracted content:

{
  "url": "https://example.com",
  "timestamp": "20231215120000",
  "archiveUrl": "https://web.archive.org/web/20231215120000/https://example.com",
  "title": "Example Domain",
  "content": "Extracted page text..."
}

Examples

Search with date range:

./scripts/wayback_search search "https://example.com" --from 20200101 --to 20201231

Search all subpages of a domain:

./scripts/wayback_search search "example.com" --match domain --status 200 -n 50

Deduplicate by content:

./scripts/wayback_search search "https://example.com" --collapse digest

Get page as it appeared on a specific date:

./scripts/wayback_search get "https://example.com" -t 20200615

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: nc9
• Source: nc9/skills
• License: MIT

Install and usage instructions live in the source repository linked above.