Wayback Machine MCP Server

Instructions

Implementation Reference

• wayback_mcp/server.py:133-159 (handler)

  Implements the core logic for retrieving an archived webpage from the Wayback Machine. Constructs the archived URL based on timestamp and optional 'original' mode, fetches the response using _fetch_text helper, and returns structured data including status, headers, and page text.

  async def get_archived_page(
  	url: str,
  	timestamp: str,
  	original: bool = False,
  ) -> Dict[str, Any]:
  	"""
  	Fetch the archived page content. Returns status, headers, and text content.
  	If `original` is True, uses the `id_` mode to minimize Wayback rewriting/banners.
  	"""
  	mode = "id_/" if original else ""
  	archived_url = f"{WAYBACK_ENDPOINT}/{timestamp}/{mode}{url}"
  	resp = await _fetch_text(archived_url)
  
  	text: Optional[str]
  	try:
  		text = resp.text
  	except Exception:
  		text = None
  
  	return {
  		"url": url,
  		"timestamp": timestamp,
  		"archived_url": archived_url,
  		"status_code": resp.status_code,
  		"headers": dict(resp.headers),
  		"text": text,
  	}

• wayback_mcp/server.py:126-132 (registration)

  Registers the 'get_archived_page' tool with the FastMCP app, specifying the name and description.

  @app.tool(
  	name="get_archived_page",
  	description=(
  		"Retrieve content of an archived webpage from the Wayback Machine "
  		"using YYYYMMDDHHMMSS timestamp. If original=true, request id_ mode."
  	),
  )