← ClaudeAtlas

design2speclisted

Convert UI designs into structured JSONC spec packages before code is written, especially for constrained platforms like extensions, dashboards, desktop shells, and mobile apps. Use for design handoff and design-to-spec workflows. Outputs specs, not implementation code.
bayeslearner/bayeslearner-skills · ★ 0 · AI & Automation · score 65
Install: claude install-skill bayeslearner/bayeslearner-skills
# Design-to-Spec Handoff Convert a reference UI design into an implementation-ready specification package that a downstream coding agent or developer can implement with minimal ambiguity. ## Why this skill exists Design-to-code is lossy, and the losses compound on specialized platforms. A coding agent looking at a mockup will reach for familiar patterns — but those patterns break when the target is a Chrome extension (CSP blocks inline scripts), a Splunk dashboard (must use SplunkUI components), or a Tauri app (UI runs in a webview with Rust IPC). This skill sits between "here's what it should look like" and "go build it." It forces a structured analysis that surfaces platform constraints, flags architectural decisions, and prevents committing to code before the constraints are understood. The structured JSONC output works as a discipline gate: the coding agent gets a spec with explicit platform rules, component mappings, and acceptance criteria rather than improvising from a screenshot. ## Core workflow 1. **Analyze the design input** — accept images, screenshots, URLs, mockups, or natural-language descriptions. Extract layout, components, color system, typography, spacing, interaction cues, responsive assumptions, and accessibility implications. 2. **Inspect project context** — if the user provides a codebase or project, infer: framework, component library, styling system, theme/token strategy, routing, build conventions, and existing design-system patterns. Classify