figma-extract

Extract Figma node trees, design tokens, and assets into a local project directory.

93.2k|10.8k|Updated Apr 28, 2026
One-click install
npx skills add https://github.com/nexu-io/open-design --skill figma-extract
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: figma-extract
Source: https://github.com/nexu-io/open-design/tree/main/plugins/_official/atoms/figma-extract
Command: npx skills add https://github.com/nexu-io/open-design --skill figma-extract

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill solves the friction of manual design-to-code handoff by automating the conversion of Figma design files into a structured, machine-readable format that developers can immediately use for implementation.

Core Features & Use Cases

  • Canonical Tree Extraction: Converts Figma node structures into a standardized JSON tree for downstream processing.
  • Token Mapping: Automatically lifts color, typography, and spacing tokens from design files.
  • Asset Export: Rasterizes and exports leaf nodes as web-ready assets (PNG, SVG, WebP).
  • Use Case: Use this to migrate a design system from Figma into a new codebase, ensuring that all design tokens and component structures are accurately captured without manual inspection.

Quick Start

Run the figma-extract skill by providing the target Figma file URL and your OAuth token to generate the project snapshot.

Frequently Asked Questions about figma-extract

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I extract design tokens from Figma?

The figma-extract skill automatically parses design tokens including colors, typography, and spacing when it processes the Figma file. These are saved into a tokens.json file within your project directory.

What happens if the Figma file has unsupported node types?

Unsupported node types are not dropped silently; they are recorded in the meta.json file under an unsupportedNodes array, allowing you to audit and address any gaps in the extraction.

Does this skill store my Figma OAuth token?

No, the daemon does not store your Figma OAuth token in SQLite or any persistent storage. The token is routed through the oauth-prompt surface specifically for the extraction process.

How can I handle component instances in the extracted tree?

The skill records componentRef links for component instances rather than treating them as duplicates. This allows downstream tools like token-map to perform accurate de-duplication.