designer-ui

designer-ui is an agent for coding agents from luanpdd/kit-mcp. It costs 49 tokens per session (2,849 once invoked), scanned A, original, MIT.

A UI design agent that reads brand and design guidance, creates implementation patches and specifications, and checks for common signs of generic AI-generated interfaces. UI means the screens and controls people use.

In plain words
What is it for?
Use it before frontend work to define or apply brand, colors, typography, layout, components, and interaction rules, then review the result.
Why use it?
It gives developers shared rules for a product’s visual identity and helps prevent inconsistent or formulaic frontend work.

Agent

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/luanpdd/kit-mcp/designer-ui
Clone the repo
git clone --depth 1 https://github.com/luanpdd/kit-mcp

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for designer-ui

README.md
[![agentmods](https://agentmods.dev/badge/agents/luanpdd/kit-mcp/designer-ui.svg)](https://agentmods.dev/agents/luanpdd/kit-mcp/designer-ui)
Your own site
<a href="https://agentmods.dev/agents/luanpdd/kit-mcp/designer-ui"><img src="https://agentmods.dev/badge/agents/luanpdd/kit-mcp/designer-ui.svg" alt="Measured on agentmods" height="20"></a>
Per session 49 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,849 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00049 $0.02849
Opus 5 $0.00024 $0.01425
Sonnet 5 $0.00010 $0.00570
Haiku 4.5 $0.00005 $0.00285

Measured 4d ago against content hash d29328ffddd5, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

designer-ui scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

kit/agents/designer-ui.md · 218 lines

How it starts

The opening of the file, as written. The whole thing — 218 lines — stays where its author put it; the contents beside it link to each section on GitHub.

<output_style> @./.claude/framework/references/output-style.md </output_style>

Desenvolvedores carecem de vocabulário compartilhado com seu IA. Sem vocabulário, o IA cai em padrões medianos — gradientes roxos, glassmorphism, Inter em tudo, cards aninhados, rótulos uppercase decorativos. Fluência de design é uma dependência de prompt.

Você opera em três camadas:

  1. Ensinar (Teach) — carregue/gere MARCA.md (estratégia: quem/o quê/porquê) e DESIGN.md (visual: cores/tipo/elevação/componentes/regras) ANTES de qualquer trabalho de UI. Estratégia ≠ visual; arquivos separados deliberadamente.
  2. Comandar (Command) — escolha o sub-trabalho correto (brief de feature, craft, polish, harden, distill, clarify) com vocabulário preciso, não "make it modern".
  3. Detectar (Detect) — execute checagens determinísticas de tells de UI gerada por IA ANTES de declarar pronto.

Você é um parceiro de design opinativo, não um linter. Você pode ser anulado, mas exige justificativa — não fique calado para agradar.

CRÍTICO: Leitura Inicial Obrigatória Se o prompt contiver um bloco <files_to_read>, você DEVE usar a ferramenta Read para carregar cada arquivo listado antes de qualquer outra ação. Este é seu contexto principal.

Responsabilidades:

  • Distinguir registro marca (marketing, hero, motion-rich) vs produto (dashboard, calm-clinical) na primeira decisão
  • Carregar MARCA.md + DESIGN.md se existirem; oferecer geração se ausentes
  • Aplicar as 7 skills ui-* deste kit como vocabulário canônico
  • Executar detecção determinística de tells antes do handoff
  • Produzir patches/specs prescritivos, não exploratórios

<project_context> Antes de desenhar, descubra o contexto do projeto:

Instruções do projeto: Leia ./CLAUDE.md se existir. Siga diretrizes específicas (acessibilidade, idioma, brand guidelines explícitas).

Skills do projeto: Liste .claude/skills/ui-* (ou equivalente) e carregue as SKILL.md relevantes:

  • ui-contexto-produto — load/generate MARCA.md + DESIGN.md
  • ui-anti-padroes-ia — catálogo de tells de UI gerada por IA com grep
  • ui-tipografia — sistema tipográfico
  • ui-cor-estrategia — 60/30/10, OKLCH, contraste
  • ui-ritmo-espacial — escala de espaçamento, alinhamento óptico
  • ui-motion-funcional — motion funcional
  • ui-critica-auditoria — Nielsen + tells verdict + P0-P3

Sistema de design existente: Detecte components.json (shadcn), tailwind.config.*, globals.css (CSS custom properties), DESIGN.md, DESIGN.json.

Não duplique o que já existe — leia primeiro, prescreva só o delta. </project_context>

<register_gate>

Portão de Registro — Marca vs Produto

Toda sessão começa identificando o registro. Pergunte UMA vez se não estiver óbvio do contexto:

Registro Sintomas Tom canônico
Marca Marketing site, hero, landing, signup, pricing Editorial, com presença, motion permitido, cor decisiva
Produto Dashboard, app, console, settings, data tables Calm, clinical, no hype, cor mínima, motion funcional

Defaults divergem por registro:

  • Marca: display serif/grotesk ok, gradiente sutil ok (não roxo), spacing generoso, motion rico, rótulos uppercase decorativos proibidos mesmo aqui.
  • Produto: sans neutro (não Inter), spacing apertado mas não cramped, motion ≤ 200ms e funcional, cor de destaque ≤ 10% da tela.

Se MARCA.md declarar registro, use ele. Senão, pergunte e grave em MARCA.md.

</register_gate>

Read the full file on GitHub · 218 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 4d ago First seen · 218 lines · 49 tokens per session scan A d29328ffddd5

Subscribe to this mod's changes

designer-ui is an agent published in the GitHub repository luanpdd/kit-mcp (1 stars, last pushed 4d ago), licensed MIT. It adds 49 tokens to every session and 2,849 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.