Doesn't this automation increase possible surface attacks on the keys themselves though? Even if using existing Qubes tools, not re-inventing the wheel, and keeping Qubes itself safe as it was before using the tool, but the automated policy can still be tricked into giving over the password though?
If true, then manual copy/paste between Qubes is supposedly more safe? Because the initiation is started from the isolated dom0 ps/2 keyboard (or USB qubed keyboard), and not initiated from within the internet exposed Qube itself.
I imagine this might be good for less important passwords, daily ones that can be annoying to type in, but also aren't too important. But regarding important passwords, perhaps use the manual method instead?
Having to use manual password copy/paste is a bit slow, takes up at the very least several seconds, if not half a minute, to open it up and navigate to find your password, and then copy/paste it over.
So it becomes a question between speed/convenience/insecure vs. slow/inconvenience/secure?
Maybe we can make a hybrid here? Like for example have a hardware key, requiring you to press it before it accepts the automated process. Or even just a popup from the isolated offline password-manager VM, before proceeding. It's not fully automated, but it's also not as intensively manual either.
Maybe the inter-VM password manager for Qubes already does something akin to requiring a single quick action from inside the offline isolated password manager VM before fulfulling the request of the online VM. If I missed it, then I apolgize, but I can't see it anywhere.
Thoughts on using a hybrid method though?