Which integrations are available for this server?

Provides tools for developing screens with Nuxt 3 using SigmaUI DS components, including listing, searching, and getting component examples.

How do I use SigmaUI DS MCP Server?

1. Click on "Install Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@SigmaUI DS MCP Server show me a usage example for the Button component" That's it! The server will respond to your query, and you can continue using it as needed. Here is a step-by-step guide with screenshots.

SigmaUI DS MCP Server

by ribsousa

Overview Schema Related Servers Score Discussions

TypeScript

Hybrid

SigmaUI DS — MCP Server

Servidor MCP (Model Context Protocol) que expõe os componentes do SigmaUI DS para assistentes de IA consultarem durante o desenvolvimento de telas Vue 3 e Nuxt 3.

Stack

Tecnologia	Papel
Bun `>= 1.1`	Runtime, bundler, test runner e gerenciador de pacotes
Hono + `@hono/mcp`	Framework HTTP + adaptador MCP
`@modelcontextprotocol/sdk` `^1.29`	SDK MCP (tools, resources, prompts)
AI SDK (`ai`, `@ai-sdk/google`, `@ai-sdk/anthropic`, `@ai-sdk/openai`)	Curadoria automática provider-agnostic
Zod `^4.0`	Validação de schemas
Prettier	Formatação de SFC (Vue) e TypeScript gerado
`@vue/compiler-sfc`	Parser oficial para extração de slots via AST

Related MCP server: Mozaic MCP Server

Instalação

git clone <url-do-repositorio>
cd sigmaui-ds-mcp-server
bun install
cp .env.example .env
bun run dev

O servidor sobe em http://localhost:3333.

Docker

docker compose up -d

O script catalog:generate precisa de acesso ao repositório pd-sigmaui-ds — rode-o localmente antes de buildar a imagem.

Scripts

Script	Descrição
`bun run dev`	Desenvolvimento com hot-reload (HTTP)
`bun run start`	Produção (HTTP)
`bun run stdio`	Integração local com IDEs/CLIs (stdio)
`bun run inspect`	MCP Inspector
`bun run catalog:generate`	Gera e curada componentes no catálogo
`bun run test`	Executa testes
`bun run test:watch`	Testes em modo watch
`bun run test:coverage`	Testes com cobertura

Variáveis de ambiente

Variável	Padrão	Descrição
`PORT`	`3333`	Porta do servidor HTTP
`ALLOWED_ORIGINS`	`http://localhost`	Origens CORS (separadas por vírgula)
`SIGMAUI_DS_PATH`	`../pd-sigmaui-ds`	Caminho para o repositório do DS
`SKIP_CURATE`	`false`	Pula curadoria IA
`CURATE_PROVIDER`	`google`	Provider: `google`, `anthropic` ou `openai`
`CURATE_MODEL`	`gemini-2.5-flash`	Modelo do provider
`GOOGLE_GENERATIVE_AI_API_KEY`	—	API key Google AI Studio
`ANTHROPIC_API_KEY`	—	API key Anthropic
`OPENAI_API_KEY`	—	API key OpenAI

Ferramentas MCP

Ferramenta	Parâmetros	Descrição
`list_components`	`category?`	Lista componentes com filtro por categoria
`get_component`	`name`	Documentação completa: props, eventos, slots, exemplo
`search_components`	`query`	Busca textual por nome, descrição ou categoria
`get_usage_example`	`name`, `framework?`	Exemplo SFC completo (vue/nuxt)
`get_design_tokens`	—	Variantes de cor e tamanhos
`get_form_pattern`	`fields`, `mode?`	Padrão de formulário com vee-validate + Zod
`suggest_components`	`description`	Sugere componentes para um caso de uso
`compare_components`	`component_a`, `component_b`	Compara dois componentes lado a lado
`get_data_table_columns`	`columns`	Gera definição de colunas para DataTable

Recursos MCP

URI	Descrição
`sigmaui://catalog`	Catálogo resumido (nome, categoria, descrição)
`sigmaui://component/{name}`	Detalhes JSON de um componente

Prompts MCP

Prompt	Descrição
`implement_screen`	Gera prompt para implementar uma tela
`create_form`	Gera prompt para formulário com vee-validate + Zod
`review_component_usage`	Revisa uso correto dos componentes
`curate_catalog_entry`	Guia curadoria de uma entrada do catálogo

O argumento framework aceita "vue" (padrão) ou "nuxt".

Integração com clientes MCP

stdio (recomendado para uso local)

claude mcp add sigmaui-ds bun /caminho/absoluto/para/sigmaui-ds-mcp-server/src/stdio.ts

HTTP (debug, deploy remoto)

bun run dev
claude mcp add sigmaui-ds --transport http http://localhost:3333/mcp

Manutenção do catálogo

O source of truth é src/data/catalog.json. O arquivo src/data/catalog.ts é derivado (gerado automaticamente).

Adicionando componentes

export GOOGLE_GENERATIVE_AI_API_KEY=sua-chave
bun run catalog:generate

O pipeline:

Extrai metadados via @vue/compiler-sfc (props, emits, slots via AST) e Storybook stories
Curada via AI SDK (provider-agnostic): corrige tipos, gera descrição/whenToUse/notes/exemplo
Formata exemplos SFC com Prettier (Vue parser, singleAttributePerLine)
Serializa catalog.ts com Prettier (TypeScript parser)
Detecta mudanças via hash SHA-256 (só reprocessa componentes alterados)

Componentes já curados são preservados — SKIP_CURATE=true não sobrescreve dados enriquecidos.

Outro provider

CURATE_PROVIDER=openai CURATE_MODEL=gpt-4.1-mini bun run catalog:generate

Testes

bun run test

Arquivo	Cobertura
`src/app.test.ts`	Health check, middleware origin validation
`src/tools/get-component.test.ts`	Markdown output, lookup, sugestões
`src/tools/search-components.test.ts`	matchesQuery, integração com catálogo
`src/prompts/templates.test.ts`	Branching vue/nuxt, formulários, curadoria
`scripts/generate-catalog.test.ts`	Utils do pipeline (toPascalCase, poolMap, hash)
`scripts/lib/utils.test.ts`	resolveComponentName, parseSubComponents
`scripts/lib/serialize.test.ts`	formatSfcExample (Prettier)
`scripts/lib/curate.test.ts`	Curadoria IA (MockLanguageModelV3)

Estrutura

sigmaui-ds-mcp-server/
├── src/
│   ├── index.ts               # Entry HTTP (Bun.serve + graceful shutdown)
│   ├── stdio.ts               # Entry stdio (stderr logging)
│   ├── app.ts                 # Hono (CORS, origin validation, transport)
│   ├── mcp.ts                 # Orquestrador McpServer
│   ├── config.ts              # Constantes (PORT, CORS)
│   ├── types.ts               # ComponentInfo interface
│   ├── data/
│   │   ├── catalog.json       # Source of truth (71 componentes)
│   │   └── catalog.ts         # Derivado (gerado por Prettier)
│   ├── tools/                 # 9 ferramentas MCP
│   ├── resources/             # Resources MCP (sigmaui://)
│   └── prompts/               # 4 prompts MCP
├── scripts/
│   ├── generate-catalog.ts    # Orquestrador do pipeline
│   └── lib/
│       ├── extractors.ts      # Parsing via @vue/compiler-sfc + Storybook
│       ├── curate.ts          # AI SDK (provider-agnostic, .nullable() schema)
│       ├── serialize.ts       # JSON.stringify + Prettier (Vue + TS)
│       ├── hashes.ts          # SHA-256 change detection
│       ├── utils.ts           # toPascalCase, poolMap, readSafe
│       └── preload.ts         # Bun plugin (mock .vue/.css)
├── Dockerfile
├── docker-compose.yml
└── .env.example

This server cannot be installed

license - not found

quality - not tested

maintenance

How are these scores calculated?

Maintenance

–Maintainers

–Response time

–Release cycle

–Releases (12mo)

Commit activity

Resources

GitHub Repository

Need Help?

Related Servers

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Servers

Components Build MCP
Developer Tools Documentation Access Code Analysis
getlokiui
A
license
B
quality
D
maintenance
Provides AI assistants with tools to grade, generate, and validate UI components against the components.build specification. Supports searching documentation, checking compliance, and generating framework-agnostic accessible components.
Last updated 2025-12-03
11
29
Apache 2.0
Mozaic MCP Server
Developer Tools Documentation Access
MerzoukeMansouri
A
license
-
quality
A
maintenance
Provides AI assistants access to the ADEO Mozaic Design System, enabling lookups of design tokens, component documentation, icons, and CSS utilities, plus generation of Vue and React component code snippets.
Last updated 2026-07-02
189
4
MIT
WordPress Design System MCPofficial
Design Tools Content Management Systems
Automattic
F
license
-
quality
F
maintenance
Provides AI assistants with access to WordPress Design System component information and design guidance.
Last updated 2026-06-08
9
GDS MCP
Design Tools Developer Tools Documentation Access
supin-kim-ilevit-com
F
license
-
quality
B
maintenance
Exposes Levit design system (GDS) metadata to AI coding tools, enabling queries about color tokens and component usage via natural language.
Last updated 2026-06-04

View all related MCP servers

Related MCP Connectors

Designesy
Design-system contract verification, scoring, and review tools for AI agents.
Hint-Services-obsidian-github-mcp
Connect AI assistants to your GitHub-hosted Obsidian vault to seamlessly access, search, and analy…
Visily UI Design
UI design from prompts, screenshots, and URLs for AI coding agents and theme tokens.

View all MCP Connectors

Latest Blog Posts

Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)
By Om-Shree-0709 on July 25, 2026.
mcp
Agent Identity
OAuth 2.1
Your AI Chatbot Just Exposed Your CEO's Salary to an Intern
By Om-Shree-0709 on July 2, 2026.
Agent Identity
MCP Security
OAuth Delegation
Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)
By Om-Shree-0709 on June 30, 2026.
Agentic Ai
Prompt Injection
WebAssembly

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/ribsousa/sigmaui-ds-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server