Wayback MCP Server

🚀 Wayback Machine MCP Server

This is a Model Context Protocol (MCP) server that provides access to the Internet Archive's Wayback Machine. It allows you to retrieve archived versions of web pages and check available snapshots of URLs.

🚀 Quick Start

This server offers access to the Internet Archive's Wayback Machine. You can use it to retrieve archived web - page versions and check available URL snapshots.

✨ Features

Tools

1. get_snapshots

   • Retrieve a list of available snapshots for a URL.
   • Parameters:
     • url (required): The URL to check.
     • from (optional): The start date in YYYYMMDD format.
     • to (optional): The end date in YYYYMMDD format.
     • limit (optional): The maximum number of snapshots to return (default: 100).
     • match_type (optional): The URL matching type to use (default: exact).
       • Options: 'exact', 'prefix', 'host', 'domain'

2. get_archived_page

   • Fetch the archived web - page content from the Wayback Machine.
   • Parameters:
     • url (required): The URL of the page to retrieve.
     • timestamp (required): The snapshot timestamp in YYYYMMDDHHMMSS format.
     • original (optional): Whether to get the original content without showing the Wayback Machine banner (default: false).

Resource Templates

1. wayback://{url}/{timestamp}
   • Access an archived web page on the Internet Archive's Wayback Machine.
   • Parameters:
     • url: The URL of the web page to retrieve.
     • timestamp: The specific timestamp (in YYYYMMDDHHMMSS format).

📦 Installation

1. Clone this repository.
2. Install dependencies: npm install
3. Build the project: npm run build
4. Add the server to the MCP settings file:

{
  "mcpServers": {
    "wayback-machine": {
      "command": "node",
      "args": ["/path/to/wayback-server/build/index.js"],
      "env": {},
      "disabled": false,
      "autoApprove": []
    }
  }
}

💻 Usage Examples

Basic Usage

Get Snapshots

use_mcp_tool(
  server_name="wayback-machine",
  tool_name="get_snapshots",
  arguments={
    "url": "example.com",
    "from": "20200101",
    "to": "20201231",
    "limit": 10
  }
)

Get Archived Page

use_mcp_tool(
  server_name="wayback-machine",
  tool_name="get_archived_page",
  arguments={
    "url": "example.com", 
    "timestamp": "20200101120000",
    "original": true
  }
)

Access Resource

access_mcp_resource(
  server_name="wayback-machine",
  uri="wayback://example.com/20200101120000"
)

📚 Documentation

API Details

This server uses the following Wayback Machine APIs:

1. Availability API: https://archive.org/wayback/available?url={url}
2. CDX Server API: https://web.archive.org/cdx/search/cdx?url={url}&output=json
3. Wayback Machine Memento API: https://web.archive.org/web/{timestamp}/{url}

📄 License

ISC