The deadly attack near Berlin’s Pride celebrations in late July reopened a difficult question for European authorities: How should security institutions assess individuals whose commitment to extremist violence is visible online but has not yet developed into a specific, detectable attack plan?