Shape a complex idea, workflow, or design around the judgment a person needs to make. Use when an explanation obscures its point or makes the user carry unnecessary complexity. Do not use for routine execution of a clear request.
Make the model easy to recognize and judge without making it less true. The right amount of detail is what this person needs to understand, correct, or act on it; brevity alone does not reduce their work.
Start from what the person is trying to accomplish and the judgment in front of them. Show the governing relationship before its machinery. Preserve the actors, ownership, sequence, and consequences that change that judgment; leave history and incidental implementation detail available for later.
Choose the form from what needs to be understood. Prose can carry a claim, a comparison can expose a trade-off, and code or a diagram can make a relationship concrete. Use visual hierarchy so the point requiring attention is easy to find. A successful layout in one conversation is not a required format for the next.
Simplification fails when it hides uncertainty, makes deferred behavior look implemented, or collapses distinct owners. Extra detail fails when the person must reconstruct the point from an inventory. Keep a distinction when removing it would change the person's understanding or decision.
dialectic develops an unsettled model through reactions; abstraction chooses how that model meets the person. Show alternatives when comparing them helps resolve the uncertainty. Otherwise present the strongest current account and what remains unresolved.
Stop when the person can make the next judgment without reconstructing the machinery behind it.
1b73add
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.