How does it decide to ask versus act?
It classifies each action by three questions — is it reversible, is it external (does it leave your accounts), and is it sensitive. Reversible, internal, non-sensitive things are handled quietly; sensitive, irreversible, or un-permissioned things are proposed and wait for your approval.
Those three questions are why an unsubscribe request is never sent for you on a standing permission — it's external and it can't be pulled back — while silencing the same sender can be: the silence is internal to your mailbox and reversible in one tap.
If something is routine and within a standing permission you've switched on, it acts under that permission and reports back. Those permissions are part of the Lonzo Assistant plan, which is coming soon; until then it proposes and waits. If it's uncertain whether something is sensitive, it surfaces it as a decision rather than guessing. The safe default always wins.
See also: What is configured autonomy? · When does the assistant act on its own?