The backup download should actually be done by the back-end as the S3 Url is usually an internal one and is not reachable from the browser.

Ideally the target download directory should be proposed (file browser) and, optionally, the associated volume information included into the download (see #80).