Dembrandt UX Skills

Bring senior UX and design-system judgment into AI-built interfaces.

Stars
★ 147
Last updated
today
License
MIT
Primary language
JavaScript

At a glance

How it runs
Agent plugin / skillCLI
Works with
Universal · cross-platformClaude CodeCodex (Partial support)
Cost
Free, no paid service needed
Setup effort
Low · running in minutes
You'll need
Node.js 18 or newerDembrandt CLI ≥ 0.39.0 for extract-designDembrandt CLI ≥ 0.23.1 for generate-ui-from-brandShell / CLINetwork accessLocal filesystemMCP Server
Typical use
A B2B SaaS design-system team wants consistent buttons, inputs, and badges across its product.
Not a fit if
  • Teams needing live-site token extraction without installing Chromium
  • Teams requiring explicitly documented support beyond Claude Code and Codex

What does this agent do, and when should you use it?

This repository packages Open Agent Skills for brand identity, design tokens, layout, components, UX principles, and technical foundations. An agent loads matching skills in response to a prompt; a broad interface-review request can invoke the full UX workflow described by the dembrandt skill. Of the 45 listed skills, 43 work after installation alone, while live-site features such as extract-design and generate-ui-from-brand require the Dembrandt engine. Install the skills with npx skills, or use the plugin marketplace in Claude Code; the README also names Cursor, Codex, and GitHub Copilot. The deliverable is a set of skills and an optional MCP interface for other agents, not a standalone hosted application.

The agent matches prompts to skills: nielsen-usability-heuristics provides a usability review, wcag-accessibility covers WCAG 2.2 AA checks, and algorithmic-color-palette derives a palette from a brand color. Other skills cover type scales, responsive layouts, forms, tab navigation, modals, loading states, and web performance. The dembrandt skill describes a six-stage UX orchestration workflow; extract-design extracts design tokens from a live site through the Dembrandt CLI or MCP, and generate-ui-from-brand turns a URL or DESIGN.md into tokens, decisions, and a UI spec. clone-website rebuilds a page from a capture and computed styles for Figma, Penpot, or code.

  1. A B2B SaaS design-system team wants consistent buttons, inputs, and badges across its product.
  2. A frontend developer wants to review contrast, keyboard access, and ARIA usage before shipping an interface.
  3. A product designer has a brand color and needs a complete UI palette derived from it.
  4. A product team is designing guided onboarding or a multi-step purchase flow.
  5. A design engineer needs tokens extracted from a live site such as Stripe and can install the Dembrandt engine.

How do you install or deploy this agent?

Install all skills from a terminal:

npx skills add dembrandt/dembrandt-skills --all

Use --global to install for every project. Claude Code users can instead add the marketplace and install its plugin inside Claude Code:

/plugin marketplace add dembrandt/dembrandt-skills
/plugin install dembrandt@dembrandt

For live-site features, install Node.js 18 or newer and add the Dembrandt MCP configuration:

{
  "mcpServers": {
    "dembrandt": {
      "command": "npx",
      "args": ["-y", "dembrandt", "mcp"]
    }
  }
}

Before the first extraction, install the browser:

npx dembrandt install-browser

How do you use this agent?

Open a UI file and ask your agent for a task in your own words, for example:

Review this interface for usability issues.

For a specific check, ask “Does this pass WCAG 2.2 AA?” To extract a live site's design system, request extraction for that site; this requires the Dembrandt CLI or MCP to be configured and the browser to be installed.

What are this agent's strengths and limitations?

Pros
  • The 45 listed skills span brand, tokens, layout, accessibility, and performance work.
  • The README says 43 skills work with the skills installation alone; only two live-site skills need the engine.
  • Skills load in response to prompts, with documented Claude Code plugin and Dembrandt MCP setup paths.
Limitations
  • The live-site extract-design and generate-ui-from-brand skills require the extra Dembrandt engine; first extraction also requires a separate Chromium installation.
  • Running the Dembrandt MCP setup requires Node.js 18 or newer and MCP configuration.
  • The README does not provide full input/output contracts for every skill, installation steps for every named platform, or skill-effect evaluation results.

How does this agent compare with similar options?

Key facts side by side with the most closely related agents.

Agent Source review Form / cost Stars Updated Language Full support on
Dembrandt UX Skills This agent Insufficient evidence Agent plugin / skillFree ★ 147 today JavaScript Claude Code
XIAS — UX/UI Agent Skills 70 · Some gaps Agent plugin / skillFree + model costs ★ 2.8k 3d ago JavaScript Claude Code
UIZZE · Better UI for Coding Agents 47 · Major gaps Agent plugin / skillFreemium ★ 66 10d ago JavaScript Codex · Claude Code
Designer Skills Pack 38 · Major gaps Agent plugin / skillFree ★ 3k 1mo ago Markdown Claude Code

How does FollowAgents rate this agent?

FollowAgents source review · FARS-2.1
Insufficient evidence
Risks and how to mitigate them
  • Not found in source: least-privilege scopingGrant only what the task needs: a dedicated account or read-only token, scoped to specific directories and repos.
  • Not found in source: confirmation before actingTurn on (or add) a confirmation step before it acts, and try it in a sandbox or test environment before real data.
  • Not found in source: data-flow disclosureWatch which external services it contacts (proxy or firewall logs) and keep sensitive data out until you know where it goes.
  • Not found in source: sensitive-data handlingUse dedicated, low-privilege, revocable API keys — never production credentials — and keep secrets out of logs.
  • Not found in source: dependency securityPin versions and run a dependency audit (npm audit, pip-audit) before installing; prefer running it in a container.
  • Not found in source: disclosed external effectsEstablish which external systems it writes to, sends to or changes, and verify with test accounts or repos before production.
  • Not found in source: rollback or recovery pathBack up first, or work on a git branch or snapshot, so its changes can be undone.
  • Not found in source: verifiable attributionInstall from the official repo or registry and check the publisher and URL to avoid look-alike packages.
  • The supplied material omits the skill contents listed in the README, so this assessment cannot determine the Agent's actual instructions or behavior.
  • The README says two skills require Dembrandt and a browser environment; the materials do not include those skill implementations or their data-handling details.
Evidence confidence: Low Reviewed Oct 22, 2026 Reviewed revision c4ad177dea5b
See the full review method →

FAQ

Which skills work immediately after installation?
The README says 43 of the 45 listed skills work from the skills installation alone; two live-site features need the Dembrandt engine.
What is required to extract design tokens from a live site?
Use the Dembrandt CLI or MCP. The docs require Node.js 18 or newer and say to run npx dembrandt install-browser before the first extraction.
Do I need API keys or a paid service?
The supplied material does not require API keys or describe a paid service; it describes Dembrandt as free and open source.
Can I use it outside Claude Code?
The README says the skills work with agents that read the Open Agent Skills format and names Claude Code, Cursor, Codex, and GitHub Copilot. Claude Code also has a plugin installation path.
View on GitHub ↗ Install ↓

Compare agents like this one

The same FARS review applied across the shortlist this agent qualifies for.

Related agents