Can you give me some use cases? Why not just use the shell history file?

The only case I can think of in which the shell file wouldn't suffice is with remote shells.

Yeah it's mostly about remote shells(I realized that I have to remove process-running-child-p). This is very helpful if you are working a lot with ssh like I do.

At the moment, I would like to focus on "infrastructure" work: bug fixes, documentation, refactoring, and so on. I believe that right now the priority is to make the code more stable and maintainable.

A while ago, I started a personal repo to add some functions to vterm: https://github.com/Sbozzolo/vterm-extra.
At the moment there are only two functions I wrote: one to quickly jump to another open vterm buffer, and the other to write more complex shell commands in a separate buffer.

I can publish this to MELPA and make it the official "contrib" repo for vterm. You could add your functions there.
We can upstream functions from vterm-extra to the main repo, once the infrastructure work is done.

What do you think?

Sorry for the late reply. However I'm not sure if it's important to put optional features into an additional repo. But I will try to add the feature and open a PR in vterm-extra.