By imehr
Extract, validate, and document complete design systems and brand identities from any URL. Uses a generator-critic validation loop with 5 quality gates to produce production-ready design tokens, component replications, and brand voice analysis.
Extract a complete design system and brand identity from a URL. Produces validated design tokens, component replications, brand voice analysis, and a comprehensive brand document.
Run validation gates against an existing brand extraction. Use when you have already extracted tokens and want to re-validate, or when validating a manually created design token file against a source URL.
Extracts design systems from URLs using Playwright, CSS analysis, and AI interpretation. Runs the full extraction pipeline (reconnaissance, token extraction, asset extraction, voice analysis, synthesis, component replication). Works with the validation-agent in a generator-critic loop.
Adversarial quality assurance agent that evaluates brand extractions against a strict 5-gate rubric. Produces machine-readable pass/fail verdicts with remediation instructions. Independent from the extraction agent to ensure unbiased evaluation. Use when validating design token extractions, component replications, or brand voice analyses.
Extract complete design systems and brand identities from URLs using Playwright, CSS analysis, and AI interpretation. Covers colour, typography, spacing, components, brand voice, and accessibility. Use when extracting design tokens from websites, reverse-engineering design systems, creating brand documentation from live sites, or building component replications from extracted tokens.
Analyse brand voice, tone, and content strategy from website text content. Extracts tone dimensions, voice characteristics, vocabulary patterns, CTA style, and language variant (AU/US/UK English). Use when analysing brand tone of voice, creating content guidelines from existing copy, or documenting brand communication style from scraped website text.
Generate W3C DTCG (Design Tokens Community Group) compliant design token files from extracted brand data. Covers colour, typography, spacing, border-radius, shadow, and motion tokens in the 2025.10 stable specification format. Use when converting extracted CSS values into structured design tokens, exporting tokens for Style Dictionary or Tokens Studio, or validating DTCG schema compliance.
Validate brand extractions by comparing replicated components against originals using a three-layer approach (pixel comparison, structural LLM analysis, token traceability). Implements Gate 5 of the validation pipeline. Use when comparing component replications to originals, performing visual regression testing on design tokens, or validating that extracted tokens accurately reproduce the source design.
Executes bash commands
Hook triggers when Bash tool is used
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Claude Code plugin that extracts complete design systems and brand identities from any URL. Uses a generator-critic validation loop with 5 quality gates to produce production-ready design tokens, component replications, and brand voice analysis.
# Add the marketplace (if not already added)
/plugin marketplace add imehr/imehr-marketplace
# Install the plugin
/plugin install brand-extractor@imehr-marketplace
/plugin install imehr/brand-extractor-plugin
The extraction scripts require Python packages and Playwright:
pip install playwright beautifulsoup4 lxml Pillow pixelmatch tinycss2 colormath Jinja2 requests
playwright install chromium
/brand-extractor:extract-brandFull 8-stage extraction pipeline with 5-gate validation:
/brand-extractor:extract-brand https://example.com
/brand-extractor:extract-brand https://example.com --brand-name "Acme Corp" --depth 3
Pipeline Stages:
/brand-extractor:validate-extractionRun the 5-gate validation pipeline against existing extractions without re-extracting:
/brand-extractor:validate-extraction ./docs/acme/extraction/
Generator-Critic Loop:
5 Quality Gates:
| Gate | Focus | Pass Threshold |
|---|---|---|
| Gate 1 | Reconnaissance | 5 blocking criteria |
| Gate 2 | Token Extraction | >= 0.85 |
| Gate 3 | Asset Extraction | >= 0.60 + logo required |
| Gate 4 | Synthesis & Interpretation | >= 0.80 |
| Gate 5 | Visual Replication (3-layer) | >= 0.83 avg |
The plugin produces:
extraction/
├── tokens.json # W3C DTCG 2025.10 design tokens
├── css-variables.css # CSS custom properties
├── design-system.html # Self-contained visual showcase
├── {brand}-logo.svg # Standalone SVG logo
├── brand-document.md # Comprehensive brand guide
├── voice-analysis.json # Tone, voice, CTA patterns
├── metadata.json # Extraction metadata
├── screenshots/ # Desktop, tablet, mobile captures
├── components/ # Replicated HTML/CSS components
└── validation/ # Gate verdict reports
| Skill | Description |
|---|---|
brand-extraction | Core extraction methodology, validation rubric, component taxonomy |
design-tokens | W3C DTCG 2025.10 token format and schema reference |
visual-validation | Gate 5 three-layer comparison (pixel, structural, traceability) |
brand-voice | Voice, tone, and content strategy analysis |
| Script | Purpose |
|---|---|
extract_tokens.py | Playwright-based token extraction (recon, tokens, assets) |
screenshot_components.py | Component screenshot capture |
compare_visual.py | Pixel comparison engine (pixelmatch) |
pixel_compare.py | Lightweight Gate 5 Layer 1 validator |
MIT
npx claudepluginhub imehr/imehr-marketplace --plugin brand-extractorAutomatically log development sessions - captures tools, decisions, accounts, and progress from conversation context
Build a comprehensive ME.MD profile that captures who you are and how you decide. Reads your CliftonStrengths report (if you have one) or runs a structured interview, then triangulates against your real prompt history across Claude Code, Codex, and OpenCode. Output is a machine-readable profile designed to be loaded into every AI session via @~/ME.MD.
Extract a complete, validated, reusable design system from any URL — patterns, relationships, motion, voice, and a self-validating shadcn/Tailwind HTML replica. Iteratively refines until the replica matches reference screenshots at >=0.85 similarity. Produces a per-brand DESIGN.md and an installable per-brand SKILL.md, indexed in a local design library at ~/.claude/design-library/.
Create multi-platform explainer videos with intelligent orchestration and learning memory
Complete workflow for creating Claude Code skills and marketplace repositories - TDD methodology, verification checklists, activation optimization, and large portfolio management. Use when creating skills, setting up marketplaces, or managing many skills.
Complete collection of battle-tested Claude Code configs from an Anthropic hackathon winner - agents, skills, hooks, and rules evolved over 10+ months of intensive daily use
Modern R development skills for Claude Code - tidyverse patterns, rlang metaprogramming, Bayesian inference, performance optimization, and more
Unity Development Toolkit - Expert agents for scripting/refactoring/optimization, script templates, and Agent Skills for Unity C# development
Complete creative writing suite with 10 specialized agents covering the full writing process: research gathering, character development, story architecture, world-building, dialogue coaching, editing/review, outlining, content strategy, believability auditing, and prose style/voice analysis. Includes genre-specific guides, templates, and quality checklists.
Comprehensive .NET development skills for modern C#, ASP.NET, MAUI, Blazor, Aspire, EF Core, Native AOT, testing, security, performance optimization, CI/CD, and cloud-native applications
20 SEO/GEO skills and 5 commands on one shared contract for keyword research, content creation, technical audits, schema markup, monitoring, quality gates, entity truth, and campaign memory.