What problem does it solve? Designing a feature's UI before code is written often produces vague ideas that never translate into buildable work. This Skill structures the explore/model phase of UI design: it reads existing descriptions and screens, proposes divergent concepts, and converges on a concrete proposal that directly seeds the build manifest's ui blocks and read-model view shapes. ## Core Features & Use Cases - Divergent concept generation: Applies How-Might-We reframing and Crazy-8 style sketching to propose 1-3 UI concepts with layouts, flows, and component breakdowns. - State-first design: Forces consideration of empty, error, and loading states for every screen, not just the happy path. - Consumer-driven output: Writes ux-proposal.md with screens, components, states, and the data views each screen needs, which seed the manifest's ui blocks and read-models' view_shape. - Use Case: During the mismAgent explore/model phase of a feature with UI, you dialogue with the skill to imagine the interface, pick one concept, and produce a proposal the FE worker later implements. ## Quick Start Ask the agent to use the ux-designer skill to propose UI concepts for the current feature based on its product brief and existing screens.
