taste

Extract design tokens and taste principles from website DOM data and screenshots.

282|21|Updated May 24, 2026
One-click install
npx skills add https://github.com/senlindesign/taste-skill --skill taste-senlindesign
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: taste
Source: https://github.com/senlindesign/taste-skill/tree/main
Command: npx skills add https://github.com/senlindesign/taste-skill --skill taste-senlindesign

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires @playwright/mcp, and includes references (resource) components.

What problem does it solve?

This Skill solves the "slop" problem in AI-generated design by moving beyond superficial token extraction to uncover the underlying "why" behind a website's visual decisions.

Core Features & Use Cases

  • Design DNA Extraction: Captures both practical tokens (colors, typography, spacing) and the opinionated trade-offs that define a site's aesthetic.
  • Multi-Tool Export: Generates ready-to-use configuration files for Cursor, Windsurf, Claude Code, and other AI coding agents.
  • Use Case: When you want to build a new dashboard that feels like Linear, use this Skill to extract their specific design principles and restraint trade-offs to guide your own development.

Quick Start

Run the taste skill on the provided URL to analyze its design system and generate the corresponding design DNA files.

Frequently Asked Questions about taste

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

FAQPage Schema
How do I reverse-engineer a website's design system to extract design tokens?

To reverse-engineer a website's design system, this Skill captures DOM data and visual screenshots to produce structured design tokens and taste principles. It extracts practical tokens like colors and spacing alongside the opinionated trade-offs defining the aesthetic.

Can I use Playwright MCP for browser-based DOM extraction to analyze web aesthetics?

Yes, Playwright MCP is required for browser-based DOM extraction and visual analysis. This dependency enables the Skill to capture the underlying visual decisions and aesthetic trade-offs of any targeted website automatically.

How do I extract design DNA from a website like Linear to guide my AI coding agent?

Extracting design DNA from sites like Linear captures their specific design principles and restraint trade-offs. This generates ready-to-use configuration files for AI coding agents like Cursor and Windsurf to guide your own dashboard development.

What is the best way to extract design principles and avoid AI-generated design slop?

The best way to avoid AI-generated design slop is to move beyond superficial token extraction by uncovering the underlying why behind visual decisions. This Skill captures both practical design tokens and the opinionated trade-offs defining a site's unique aesthetic.

Does this design analysis tool export configuration files for Claude Code and other AI agents?

Yes, this design analysis tool exports multi-tool configuration files for Claude Code, Cursor, and Windsurf. These generated files provide ready-to-use design guidance for AI coding agents to replicate a specific aesthetic.

Why do I need Playwright MCP to analyze a website's visual design?

You need Playwright MCP because it provides the browser-based DOM extraction and visual analysis required to reverse-engineer a website. This environment setup allows the Skill to capture both structural data and visual screenshots for accurate design token generation.