I have looked into this question over the years, but as a newb, without fully understanding whitepapers, I have never gotten a satisfying answer.
I am specifically wondering about the ability (not probability) of a threat actor compromising the main key and gaining access to ALL your accounts (thereby making it so much easier for them to cause trouble).
Is there a manager that takes this into consideration despite it’s irregularity and designed the service to mitigate this threat? Or does the act of mitigating this threat make the service cumbersome, in some way, not usable?
The ultimate question is if a person is targeted by a highly intelligent threat actor, would using a password manager be less secure than creating random pwds manually for every account?