AGENT SKILL

design-to-code

Pixel-perfect Figma to React conversion using coderio. Generates production-ready code (TypeScript, Vite, TailwindCSS V4) with high visual fidelity. Features robust error handling, checkpoint recovery, and streamlined execution via helper script.

License 许可
Apache-2.0
最近更新
2026-11-14
Stars
★ 3.3w
Forks
3.9k

01 前言 Front Matter

name
design-to-code
description
Pixel-perfect Figma to React conversion using coderio. Generates production-ready code (TypeScript, Vite, TailwindCSS V4) with high visual fidelity. Features robust error handling, checkpoint recovery, and streamlined execution via helper script.

02 资源文件 3

03 SKILL.md 正文

Design to Code

High-fidelity UI restoration from Figma designs to production-ready React + TypeScript components. This SKILL uses a robust helper script to minimize manual errors and ensure pixel-perfect results.

Prerequisites

  1. Figma API Token: Get from Figma → Settings → Personal Access Tokens
  2. Node.js: Version 18+
  3. coderio: Installed in scripts/ folder (handled by Setup phase)

Workflow Overview

Phase 0: SETUP    → Create helper script and script environment
Phase 1: PROTOCOL → Generate design protocol (Structure & Props)
Phase 2: CODE     → Generate components and assets

Phase 0: Setup

Step 0.1: Initialize Helper Script

User Action: Run these commands to create the execution helper and isolate its dependencies.

mkdir -p scripts

# 1. Copy script files
# Note: Ensure you have the 'skills/design-to-code/scripts' directory available
cp skills/design-to-code/scripts/package.json scripts/package.json
cp skills/design-to-code/scripts/coderio-skill.mjs scripts/coderio-skill.mjs

# 2. Install coderio in scripts directory (adjust version if needed)
cd scripts && pnpm install && cd ..

Step 0.2: Scaffold Project (Optional)

If starting a new project:

  1. Run: node scripts/coderio-skill.mjs scaffold-prompt "MyApp"
  2. AI Task: Follow the instructions output by the command to create files.

Phase 1: Protocol Generation

Step 1.1: Fetch Data

# Replace with your URL and Token
node scripts/coderio-skill.mjs fetch-figma "https://figma.com/file/..." "figd_..."

Verify: process/thumbnail.png should exist.

Step 1.2: Generate Structure

  1. Generate Prompt:

    node scripts/coderio-skill.mjs structure-prompt > scripts/structure-prompt.md
    
  2. AI Task (Structure):

    • ATTACH: process/thumbnail.png (MANDATORY)
    • READ: scripts/structure-prompt.md
    • INSTRUCTION: "Generate the component structure JSON based on the prompt and the attached thumbnail. Focus on visual grouping. Use text content to name components accurately (e.g. 'SafeProducts', not 'FAQ')."
    • SAVE: Paste the JSON result into scripts/structure-output.json.
  3. Process Result:

    node scripts/coderio-skill.mjs save-structure
    

Step 1.3: Extract Props (Iterative)

  1. List Components:

    node scripts/coderio-skill.mjs list-components
    
  2. For EACH component in the list:

    a. Generate Prompt:

    node scripts/coderio-skill.mjs props-prompt "ComponentName" > scripts/current-props-prompt.md
    

    b. AI Task (Props):

    • ATTACH: process/thumbnail.png (MANDATORY)
    • READ: scripts/current-props-prompt.md
    • INSTRUCTION: "Extract props and state data. Be pixel-perfect with text and image paths."
    • SAVE: Paste the JSON result into scripts/ComponentName-props.json.

    c. Save & Validate:

    node scripts/coderio-skill.mjs save-props "ComponentName"
    # If this fails, re-do step 'b' with better attention to the thumbnail
    

Phase 2: Code Generation

Step 2.1: Plan Tasks

node scripts/coderio-skill.mjs list-gen-tasks

This outputs a list of tasks with indices (0, 1, 2...).

Step 2.2: Generate Components (Iterative)

For EACH task index (starting from 0):

  1. Generate Prompt:

    node scripts/coderio-skill.mjs code-prompt 0 > scripts/code-prompt.md
    # Replace '0' with current task index
    
  2. AI Task (Code):

    • ATTACH: process/thumbnail.png (MANDATORY)
    • READ: scripts/code-prompt.md
    • INSTRUCTION: "Generate the React component code. Match the thumbnail EXACTLY. Use STRICT text content from input data, do not hallucinate."
    • SAVE: Paste the code block into scripts/code-output.txt.
  3. Save Code:

    node scripts/coderio-skill.mjs save-code 0
    # Replace '0' with current task index
    

Step 2.3: Final Integration

Inject the root component into App.tsx. Use the path found in the last task of Phase 2.1.


Troubleshooting

  • "Props validation failed": The AI generated empty props. Check if process/thumbnail.png was attached and visible to the AI. Retry the props generation step.
  • "Module not found": Ensure node scripts/coderio-skill.mjs save-code was run for the child component before the parent component. Phase 2 must be done in order (0, 1, 2...).
  • "Visuals don't match": Did you attach the thumbnail? The AI relies on it for spacing and layout nuances not present in the raw data.

04 同仓库更多 Skills

davila7
davila7/claude-code-templates

wealthville-pool-scoring

Score DeFi liquidity pools before providing liquidity. Returns a 0-100 pool score and an Enter/Hold/Exit/Reduce/Avoid verdict with calibrated confidence for ~68,800 Solana pools and 575 EVM pools, plus a public miss-inclusive track record. Use when asked whether a pool is worth LPing into, which pools look best right now, or how accurate the scores have actually been.

Skill与智能体开发
★ 3.3w
davila7
davila7/claude-code-templates

technical-debt-auditor

Scans a codebase and produces a quantified technical debt report with a prioritized remediation plan. Analyzes dead code, TODO/FIXME/HACK markers, dependency staleness, complexity hotspots, code duplication signals, test coverage gaps, unused exports, and outdated patterns. Use this skill when the user asks to audit their codebase for tech debt, wants a health check of their code quality, says things like "where should I focus refactoring", "find code smells", "audit my codebase", "what technical debt do we have", "prioritize cleanup work", or asks about code quality metrics. Also trigger when the user mentions "code rot", "legacy cleanup", "refactoring priorities", or wants to understand the overall health of a project before a major effort. Trigger even if the user doesn't say "technical debt" explicitly — any request to systematically assess code quality and identify improvement priorities belongs here.

Skill与智能体开发
★ 3.3w
davila7
davila7/claude-code-templates

readonly-sql-guard

Make an agent's database access provably read-only. Use when an agent runs caller-supplied SQL against a production/ERP/analytics database (a "query" MCP tool, text-to-SQL, an admin console), or when reviewing DB-access code for write/DDL/side-effect/transaction-escape risks. Checks what a statement CALLS, not just its shape.

数据分析师后端与接口Skill与智能体开发
★ 3.3w
davila7
davila7/claude-code-templates

web-artifacts-builder

Suite of tools for creating elaborate, multi-component claude.ai HTML artifacts using modern frontend web technologies (React, Tailwind CSS, shadcn/ui). Use for complex artifacts requiring state management, routing, or shadcn/ui components - not for simple single-file HTML/JSX artifacts.

前端开发
★ 3.3w
davila7
davila7/claude-code-templates

webapp-testing

Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.

测试与质量前端开发
★ 3.3w
davila7
davila7/claude-code-templates

discernment-nudge

After you give a substantive answer or draft that the user may act on — advice or recommendations, drafted artifacts such as goals, plans, pitches, proposals, or emails, estimates or projections, analysis or interpretation of data, factual claims they may rely on, or a multi-step argument — invoke this skill BEFORE finalizing your reply and then, if it applies, append 2-3 short follow-up questions, each tied to something specific in what you just produced, that help the user check key facts, probe the reasoning or assumptions, and notice missing context. Do this at most once per conversation. Skip it when the user asked a trivial how-to or simple lookup, wants a purely educational explanation, asked you only to format, convert, or assemble a file from content they provided, is writing code they will run, is doing creative writing or casual chat, or already asked you to double-check, cite, or review — the skill file explains these boundaries and the exact output format.

内容创作者写作与文案Skill与智能体开发
★ 3.3w