Put one settled thing in the user's head, plainly. Use when the user does not hold something already settled.
Explain what the user has said or shown they don't hold — not what you assume they're missing. Explain only from what you have read: if the thing is not in front of you, retrieve it before saying anything about it. If it cannot be retrieved, name what's missing, ask the user for it, and wait.
State the thing in one plain line — no jargon, no preamble, no build-up. Stop there. Name the two or three directions they could pull next, and wait.
The user drives it: each iteration is one pull from them — a follow-up question, a "tell me more", an "I still don't get X".
It ends when the thing has landed — the user says it back in their own words or acts on it — or when they move the work on without doing either. Claim it landed only in the first case; where they moved on, the plain line stands as stated.
The user holds it and you're the one missing it → ask a direct question. The user holds it but cannot yet put it into words → use elicit skill. The user wants a verdict on it, not an understanding of it → use judge skill.
595eda0
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.