Rule creation intelligence
under consideration
D
Doug Thomas
It would be nice if AutoElevate would be able to indicate that there was a possibly dangerous rule that was created. For example, if someone created a rule to allow the cmd prompt at the global level. That might be something that could be "flagged" and filtered by so that someone could check into it and change it if necessary.
D
Daniel Rivera
updated the status to
under consideration
D
Doug Thomas
It would also be amazing if you could indicate how frequently a rule was being used so we could identify stale rules. Again, maybe this is already part of another request somewhere.
D
Daniel Rivera
Doug Thomas: Good question on rule usage. That isn't surfaced in the portal today, but it's already captured and Planned: see "Detailed Rule Activity" on this board, which covers a use count along with a "date last used" column, aimed at exactly the stale-rule cleanup you're describing. Worth adding your vote there so you pick up those updates directly.
While I'm here, an update on your original post: flagging risky rules has since become part of a broader set of rule-intelligence work our product team is exploring. The direction is guidance at rule creation time — surfacing when a rule is broader in scope than it needs to be, recommending stronger matching criteria, and flagging when a similar rule already exists — plus background analysis across your whole rule set to catch redundancy and overly permissive rules after the fact. Your cmd-prompt-at-global-level example is exactly the risk case that work is aimed at.
That side is still in discovery, so no timeline yet, but we'll post here as it progresses.
D
Daniel Rivera
Thank you for feedback, Doug. This is something we are interested in looking into.