跳到技能内容
词映 AI导航
发现工具选型指南Skills 技能AI 快讯AI图片AI视频AI文档AI办公AI助手AI搜索AI编程AI音频AI设计AI智能体模型与开发AI学习模型评测AI提示词内容检测AI金融AI 3D扫描与识别其他工具跨境电商 AI
词映 AI导航
工具库Skills 技能
我的收藏
词映 AI导航
发现工具选型指南Skills 技能AI 快讯AI图片AI视频AI文档AI办公AI助手AI搜索AI编程AI音频AI设计AI智能体模型与开发AI学习模型评测AI提示词内容检测AI金融AI 3D扫描与识别其他工具跨境电商 AI
Skills directory/Jason904/ui-skill-lab/visual-to-spec

Codex Visual to Spec for Screenshot-Based Frontend Contracts

Turn a finalized screenshot into layout documentation, a component tree, and design tokens with evidence and confidence annotations.

中文

Install in Codex

Codex with image input, a readable screenshot, and a writable project directory.

  1. Copy the pinned `.codex/skills/visual-to-spec/` directory and its references into the project's `.agents/skills/`.
  2. Open the project, attach the screenshot, and select `$visual-to-spec`.

Example task

Supply a mock mobile to-do screen with a heading, two items, and a bottom button. Request bounds for all visible elements.

What to check

The `03_visual_spec/` output traces components back to their screenshot positions and flags uncertain fonts and interactions.

Original SKILL.md on GitHub · Updated 2026-11-13

Frequently asked question

Does a screenshot establish what a button does?

Treat the visible state as evidence and record unshown behavior for manual review.

Browse Codex skills in the Chinese directory
词映 AI 导航Skills 技能
工具分类全部工具已本地化技能选型指南AI 快讯