What problem does it solve?
It prevents agents from making the wrong edit when a UI request is missing key context (where the change should happen, what the current state is, and what the desired state should be).
Core Features & Use Cases
- Three-slot clarification: Collects location, current behavior, and desired behavior before editing to ensure the change lands on the correct element.
- Smallest-change enforcement: Updates only the minimal lines needed to achieve the visible tweak, avoiding refactors or unrelated improvements.
- Scope guardrails: Rejects or redirects requests that are actually new features, multi-component changes, or bug investigations where behavior is broken.
- Example use case: A user says a button “looks off” and wants it “moved and aligned”; the skill first asks for the exact element and what it currently looks/does, then applies only the necessary styling or layout adjustment.
Quick Start
Ask: “Tweak this UI—here’s the element location, what it looks/does right now, and what it should look/do instead.”