HomeProjectsArticlesPK BoardAbout
Explore Projects

drawio-skill

Active
GitHub Python MIT

Description

An Agent Skill turning natural language, codebases, Terraform or SQL into maintainable .drawio architecture diagrams, with layout-preserving sync, architecture tests and a built-in MCP server.

Key Features

  • Multi-source import — 13 extractors cover Python/JS/Go/Rust import graphs, Terraform/K8s/docker-compose resources, SQL DDL, OpenAPI and CI pipelines
  • Layout-preserving incremental sync — diagramctl sync updates only changed nodes and relations while tuned coordinates, styles and annotations survive
  • Diagram-as-Test — YAML/JSON rules check tier violations, cycles, orphans and more, with an official GitHub Action for CI
  • Official shape search — resolves among 10000+ official draw.io shapes and 321 AI/LLM brand logos instead of blank-box guesses
  • Built-in MCP server — diagramctl_mcp.py exposes nine tools, stdlib-only and offline, for Claude Desktop, Cursor and other hosts
  • Self-check with auto-fix — reads its own PNG output and fixes overlaps, clipped labels and stacked edges in up to two rounds, plus a five-round feedback loop

Use Cases

💡 Generate a microservices e-commerce architecture diagram from a single natural-language prompt and export it to PNG or SVG for review decks
💡 Visualize existing Terraform or Kubernetes manifests with official cloud icons to audit what is actually deployed
💡 Enforce architecture contracts on every pull request in CI to catch tier violations and structural regressions
💡 Extract and auto-lay out module import graphs from a codebase so new engineers grasp the structure quickly

Strengths & Limitations

✅ Strengths

  • • Actively maintained, recent updates
  • • High community interest (9.2k stars)
  • • Permissive open-source license (MIT)
  • • Responsive to issues, low backlog

Tags

diagrams drawio architecture visualization mcp python

Categories

⚡ Agent Tools 🔌 MCP Tools

Quick Start

# Install the draw.io desktop CLI (macOS, version 30+ recommended)
brew install --cask drawio
# Install the skill for any agent
npx skills add Agents365-ai/365-skills -g
# Or clone manually into your skills folder
git clone https://github.com/Agents365-ai/drawio-skill.git ~/.claude/skills/drawio-skill
Visit GitHub Visit Website

Related Projects

Awesome LLM Skills

1.7k · Python
Normal C

A curated list of awesome LLM and AI Agent Skills, resources and tools for customising AI Agent workflows. Works with Claude Code, Codex, Gemini CLI and custom agents.

agenttoolsmcp +2
  • · Curated collection of LLM skills for Claude Code, Codex, Gemini CLI, and more
  • · Covers document processing (docx, pdf, pptx, xlsx), development tools, and research workflows
  • · Cross-platform compatibility with Anthropic, OpenAI, Google, and Alibaba AI tools
View on GitHub

Excel MCP Server

4.2k · Python
Stale B

Model Context Protocol server for Excel file manipulation, enabling AI agents to read, create and modify spreadsheets

mcpexcelspreadsheet +2
  • · Create, read, update, and manage Excel workbooks and worksheets without Microsoft Excel installed
  • · Full data manipulation: formulas, formatting, charts, pivot tables, and Excel table operations
  • · Data validation for ranges, formulas, and data integrity with conditional formatting support
View on GitHub

Flint

4.2k · TypeScript
Active A

A visualization language that lets AI agents compile simple, human-editable chart specs into expressive charts across Vega-Lite, ECharts, Chart.js, Plotly and native Excel.

visualizationchartsmcp +2
  • · Semantic chart specs - Describe what each field means with 70+ semantic types instead of hardcoding scales, axes and spacing
  • · Multiple backends - Compile one input to Vega-Lite, ECharts, Chart.js, Plotly or native Excel charts
  • · Formal visual themes - Ten built-in presets such as NYT, Economist and Swiss, with preset inheritance and brand overrides
View on GitHub