I think I've understood the source of the confusion: there are two local repos that this software interacts with. If --repo isn't set, it reads the origin remote URL from the repo in the local directory (I'll refer to this as the installed repo), then it clones a repo, presumably in a temp directory. The latter I will refer to as the cloned repo.

I didn't intend to interact with the cloned repo's remote's name, but I now presume that I have. However, I believe it doesn't matter what remote name it uses, as long as it handles the deployment properly, so I think my code is okay there.

My intent was to have the --remote option change which remote URL is read from the installed repo, and I think my code does that.