Wayback

You can run and interact with the Wayback MCP Server to list snapshots, fetch archived pages, and search items from the Internet Archive. This MCP exposes convenient endpoints and a local runtime you can integrate with your MCP client workflow to access archived web content programmatically.

How to use

Use a compatible MCP client to call the available tools and fetch archived content from the Wayback Machine. You can list snapshots for a URL, retrieve a specific archived page, and search items in the archive. The server exposes a resource fetch method that returns archived page content for a given URL and timestamp.

How to install

Prerequisites: Python 3.12 or newer, and a shell on your system. You may also use a runtime tool like uvx to run the MCP server without a local Python install.

Install using Python’s package manager and run the server directly with Python, or run the CLI if you prefer the packaged entry point. You can also run using uvx if you want to pin a Python version and package version.

# Create a virtual environment
python -m venv .venv
source .venv/bin/activate

# Install the MCP package
pip install wayback-mcp

# Run the MCP server via Python module
python -m wayback_mcp.server

# Or run the CLI directly (provided by the package)
wayback-mcp

# Run with uvx (no install)
uvx wayback-mcp
# Pin Python version when using uvx
uvx --python 3.12 wayback-mcp
# Pin package version when using uvx
uvx --from wayback-mcp==0.1.1 wayback-mcp

Configuration and usage notes

Configure your MCP client to point to the Wayback MCP Server using a stdio configuration. You can run the server locally and reference it from your client as shown in the examples.

{
  "mcpServers": {
    "wayback-machine": {
      "command": "wayback-mcp",
      "args": [],
      "env": {},
      "disabled": false,
      "autoApprove": []
    }
  }
}

Usage examples

The following tool calls demonstrate how you can retrieve data from the Wayback MCP Server using your MCP client. Replace placeholders with your target URL and parameters.

# List snapshots for a URL
get_snapshots(url="example.com", from="20200101", to="20201231", limit=10)

# Fetch an archived page
get_archived_page(url="example.com", timestamp="20200101120000", original=true)

# Search items in the archive
search_items(
  query="title:(Wayback) AND creator:(Internet Archive)",
  mediatype="texts",
  fields=["identifier","title","creator","mediatype","publicdate"],
  sort=["publicdate desc"],
  rows=20,
  page=1
)

# Fetch a resource via the dedicated Wayback URL format
wayback://example.com/20200101120000

Notes on endpoints and data sources

Snapshot data is retrieved via the CDX API, and page content can be fetched from Wayback with the appropriate timestamp. Advanced item search can query the Archive.org services for metadata and content.

Available tools

get_snapshots

Lists snapshots for a given URL within a specified date range using the Wayback CDX API.

get_archived_page

Fetches the archived page content for a URL at a specific timestamp, returning the original or cached content as requested.

search_items

Searches Internet Archive items with metadata filters such as mediatype, fields to return, sort order, and pagination.

wayback_resource

Access the archived page content through the wayback://{url}/{timestamp} resource path.