convez.ai
MCP ServerDesign & UIIntermediário
isatimur

Excalidraw Skill Pack

isatimur/excalidraw-skill-pack

Diagramas que argumentam, não caixas que rotulam

Ver no GitHub
InicialPluginScore · de 0 a 100

MCP server que habilita agentes de IA a gerar diagramas de alta qualidade no Excalidraw com uma metodologia opinionada focada em clareza visual e argumentação didática. Integra-se nativamente com Claude Code, Cursor, Codex, Gemini CLI e qualquer cliente MCP compatível, oferecendo múltiplos temas profissionais e um renderer agnóstico.

Instalação

$npx convez-ai isatimur/excalidraw-skill-pack

1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.

Métricas

Estrelas
4
Forks
0
Acompanhando
0
Saúde
71%
Último commit
há 2 dias
Versão
@excalidraw-skill-pack/core@0.2.3
Issues abertas
1
Contribuidores
4

Casos de uso

  • Gerar diagramas arquiteturais e fluxogramas de negócio com agentes IA
  • Documentação técnica visual automatizada para projetos e APIs
  • Criação de materiais educacionais e apresentações visuais programáticas

Veredito

Ferramental maduro e bem fundamentado que eleva a qualidade visual de diagramas gerados por IA através de metodologia e temas profissionais. Adequado para equipes que precisam produzir documentação visual de alto padrão sem intervenção manual.

README

A tradução para o português deste README ainda está na fila.

Install

ForCommand
Claude Codenpx @excalidraw-skill-pack/install claude-code
Cursornpx @excalidraw-skill-pack/install cursor
GitHub Copilotnpx @excalidraw-skill-pack/install copilot
Codexnpx @excalidraw-skill-pack/install codex
Gemini CLInpx @excalidraw-skill-pack/install gemini-cli
Any MCP agentnpx @excalidraw-skill-pack/mcp-server
Renderer only (Node)npx @excalidraw-skill-pack/render diagram.excalidraw --theme stripe-press
Renderer only (Python)pipx install excalidraw-skill-pack-render && excalidraw-render diagram.excalidraw --theme stripe-press

Why this, not just another Excalidraw MCP?

Excalidraw's official MCP and most community servers solve plumbing: get an agent to emit valid JSON and render it. They produce correct boxes-and-arrows.

This pack solves taste: an opinionated methodology — the isomorphism test, evidence artifacts, multi-zoom architecture, container discipline, one-accent-per-diagram — that decides what to draw and why, so the output is a visual argument that teaches instead of a labeled grid. That's the part a weekend of MCP code can't copy, and it's why the book gallery looks the way it does.

Plumbing is a commodity now. Quality isn't.

Themes

Five themes ship in v0.1. Authoring a new theme is 20 lines of JSON + npm publish:

npx @excalidraw-skill-pack/create-theme my-brand
cd theme-my-brand && npm publish --access public

Browse the theme registry →

Published packages (v0.1)

ChannelPackagePurpose
npm@excalidraw-skill-pack/coreMethodology + bundled default-sketchy theme + schemas
npm@excalidraw-skill-pack/renderNode renderer (excalidraw-render CLI)
npm@excalidraw-skill-pack/mcp-serverstdio MCP server with 5 tools
npm@excalidraw-skill-pack/installOne-command adapter installer
npm@excalidraw-skill-pack/create-themeScaffolder for new theme packages
npm@excalidraw-skill-pack/theme-stripe-pressEditorial / book-grade
npm@excalidraw-skill-pack/theme-notionRounded, off-white
npm@excalidraw-skill-pack/theme-whiteboardLow-fi, bright, sketchy
npm@excalidraw-skill-pack/theme-darkInverted contrast
PyPIexcalidraw-skill-pack-renderPython renderer (excalidraw-render CLI)

Proof

This skill drew 77 diagrams for the published technical book From Copilot to Colleague — argument spines, chapter openers, concept figures, and inline explainers, all from the methodology in this repo. No other Excalidraw generator can point at a corpus like that. A curated set — the book's argument spine plus all ten chapter openers — ships as source .excalidraw files in examples/book; the full 77 are live across fromcopilottocolleague.com.

Methodology

Diagrams are arguments. The shape should BE the meaning.

  • Isomorphism Test: would the structure alone communicate the concept?
  • Evidence artifacts: real code snippets, actual API names, concrete formats — not placeholder text.
  • One accent per diagram. Two means a competing argument; split it.

Read the full methodology (it's also what the AI agent reads).

License

MIT. Contributions welcome — see CONTRIBUTING.md.

Ver o original no GitHub

Bom para

Relacionados