What problem does it solve? Converting Figma design mockups into runnable front-end code by hand is slow and error-prone. This Skill lets an AI agent read a Figma design URL, decompose layers by naming prefixes, slice images, generate code, and run per-block visual comparisons automatically. ## Core Features & Use Cases - Figma-to-code generation: Parses Figma nodes through the REST API (no MCP plugin or OAuth) and produces React H5, React Native, or Taro/xtaro cross-platform code with configurable style formats (scss, less, css-modules, tailwind, inline, styled-components, nativewind). - Prefix-driven layer protocol: Designers name layers with prefixes like sub-, img-, bg-, btn-, fixed-, list- to control block splitting, image slicing, and layout semantics. - Rule-based quality gates: 24 rules (R01-R24) plus a check-rules.mjs validator enforce fidelity on positioning, colors, padding, flex direction, sizing, and image slicing, with per-block visual screenshot comparison against the design. - Use Case: Paste a Figma design link into Claude Code or Codex after running npx @double-coding/pixel-print init; the agent slices assets, dispatches block generation, and delivers code plus visual diff screenshots. ## Quick Start Ask the agent to convert this Figma design into code by pasting the design URL, for example: turn https://figma.com/design/xxx?node-id=1-2 into a React page.