Internally it's taking this URL and downloading the file, building the path with

file:///

(edited)

This part is confusing, http link should work, how are you ending up with file:// links ? do you want to provide the full agent log with the error ?