3 incidents
Risk: Agent takes destructive or unapproved actions Control: Tool allow-list Outcome: Not mentioned
HeldFailedGot aroundMissingNot mentionedNot read yetAttacker's ownOrder: model · harness · application · target
Risk: Agent takes destructive or unapproved actions Control: Tool allow-list Outcome: Not mentioned
HeldFailedGot aroundMissingNot mentionedNot read yetAttacker's ownOrder: model · harness · application · target