design extractor

    1

    Clone any website's design into a reusable DESIGN.md — colors, typography, spacing, components, and a ready-to-use rebuild guide.

    Free

    1 useSecurity scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    design extractor

    Example session with this skill installed

    Extract the design system from linear.app and create a design spec for a new project.

    • Read your context and instructions
    • Compiled the design extractor

    I have captured and analyzed the visual design system for linear.app. The site uses a sophisticated, high-contrast "dark mode" aesthetic with precise typography and subtle elevation.

    DESIGN.md — Linear (linear.app)

    1. Overview

    A minimalist, high-performance visual language characterized by a deep charcoal palette, crisp sans-serif typography, and subtle border-based elevation rather than heavy shadows. The vibe is professional, focused, and utilitarian.

    2. Color Palette

    TokenValueRole
    background#0b0c0dPrimary page background
    surface#161718Card and navigation background
    primary#5e6ad2Call-to-action buttons and accents
    text-main#eeeeeePrimary body and heading text
    text-muted#8a8f98Secondary labels and metadata
    border#2e3035Thin dividers and component outlines

    3. Typography

    • Primary Font: Inter (System sans-serif fallback)
    • Scale: 12px (tiny), 14px (base), 16px (ui), 24px (heading), 48px (hero)
    • Weights: 400 (regular), 500 (medium), 600 (semibold)

    4. Implementation Guide

    // tailwind.config.js snippet
    module.exports = {
      theme: {
        extend: {
          colors: {
            linear: {
              bg: '#0b0c0d',
              surface: '#161718',
              brand: '#5e6ad2',
              border: 'rgba(255, 255, 255, 0.08)',
            }
          },
          fontFamily: {
            sans: ['Inter', 'system-ui', 'sans-serif'],
          },
          borderRadius: {
            'linear': '6px',
          }
        }
      }
    }
    

    Next steps

    1. Review the full-page screenshots in the /capture folder to see the component spacing in action.
    2. Run the provided Tailwind config to initialize your new project styles.
    3. Reference the DESIGN.md for specific button hover state transitions.
    DESIGN.md

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Stop guessing hex codes or manually inspecting elements to replicate a site's aesthetic. This skill automates the extraction of full design systems from any live URL, turning visual styles into a structured DESIGN.md specification. It bridges the gap between seeing a design and having the technical tokens needed to rebuild it.

    What it does

    • Automated capture runs a headless browser to snapshot layouts and extract computed styles from any domain.
    • Token extraction pulls exact hex values, font stacks, spacing scales, and border radii directly from the DOM.
    • Visual synthesis analyzes screenshots to define the "vibe" and component patterns that raw CSS data often misses.
    • Spec generation writes a comprehensive DESIGN.md including typography scales and elevation tokens.
    • Code generation outputs ready-to-use CSS custom properties and Tailwind CSS configuration snippets.

    How it works

    1. Capture phase executes a Python script to render the homepage and key sub-pages for visual and data analysis.
    2. Analysis phase correlates screenshots with tokens.json to identify primary actions, surface colors, and type hierarchies.
    3. Synthesis phase compiles the findings into a standardized markdown design document with implementation guides.

    Frameworks & tools

    Uses Python 3.9+, Playwright (Chromium), and CSS computed properties. Generates output compatible with Tailwind CSS and standard CSS variables.

    Why this beats prompting it yourself

    Standard prompts rely on the LLM's training data, which is often outdated or vague regarding specific site versions. This skill uses live execution to pull current, real-world computed values, ensuring your technical specs match the latest deployment of a target site.

    Use cases

    • Replicating a specific visual style for a new internal tool or dashboard.
    • Creating a design system for a client based on their existing legacy website.
    • Reverse-engineering the typography and spacing of high-end SaaS landing pages.
    • Building a Tailwind config that matches a reference site's branding exactly.

    Known limitations

    Heavy bot protection or login walls may restrict capture to a basic HTML/CSS scrape. Does not extract copyrighted assets or trademarked logos.

    Requires Python 3.9+ and Playwright. Blocked sites fall back to basic HTML scraping. Does not extract logos or proprietary images.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    1 use

    Times used to date

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 1 month ago

    • Free with an Agensi account

    Requires

    Python
    macOS

    Permissions

    Terminal / Shell
    Browser
    Network Access
    Listed1 month ago

    What's inside

    Frequently Asked Questions

    This skill uses a live headless browser to extract real-time computed styles from a URL, whereas manual prompting relies on an AI's potentially outdated training data or visual estimation.

    You will receive the complete design extractor tool, documentation on how to run the capture scripts, and templates for the DESIGN.md and Tailwind configuration outputs.

    It is designed to run in environments supporting Python 3.9+ and Playwright, making it compatible with most local developer setups and advanced AI agents that have terminal or script execution capabilities.

    Yes, the skill is specifically built to bridge the gap between CSS and modern frameworks by generating ready-to-use CSS variables and Tailwind CSS configuration snippets.

    The skill extracts technical design tokens like colors, fonts, and spacing; however, it does not download copyrighted images, logos, or proprietary assets.

    The tool performs a 'best-effort' scrape by analyzing the rendered DOM, though sites with extreme bot protection or login requirements may limit the depth of data captured.

    New in Frontend & Web Apps