For example, it can capture the entire DOM and send it off, including the contents of input fields that have not been submitted.
That DOM capture can be replayed on a browser to show what the user sees. So what’s the difference?