Is there a way to download a file with this MCP?

[JBetz]

For example, what happens if you use navigate to go to: https://file-examples.com/wp-content/storage/2017/10/file-sample_150kB.pdf

Does it return the file bytes in the response? Or is there a separate mechanism for downloading the file?

[Labeeb2339]

navigate_page itself does not return the raw PDF bytes. It performs the navigation and returns page state.

Current chrome-devtools-mcp has a separate network-response path:

1. Navigate to the URL.
2. Call list_network_requests, optionally filtering for document.
3. Take the PDF request's reqid.
4. Call get_network_request with that reqid and a responseFilePath.

For example:

get_network_request({
  reqid: 12,
  responseFilePath: C:/tmp/file.network-response
})

With responseFilePath, the implementation calls response.buffer() and saves the raw bytes. Without it, the response is formatted for the MCP result rather than being a reliable binary-download mechanism. You can rename the saved file to .pdf after checking the request status and Content-Type.

Fetch it soon after navigation because Chrome may no longer retain an older response body.

This is based on the current get_network_request schema and its binary response-file handling.

[JBetz]

A little clunky but that should work. Thanks!