Agents365-ai/drawio-skill

Agent skill that turns natural language, code, Terraform/K8s, SQL, OpenAPI, AsyncAPI, Protobuf and GraphQL sources into editable, tested draw.io architecture diagrams: incremental sync, multi-view projection, drift diff, CI architecture tests, whiteboard derasterize, interactive HTML/PPTX/Mermaid exports.

What it solves

This project automates the creation and maintenance of professional architecture diagrams. It bridges the gap between natural language prompts or existing codebases and editable .drawio files, solving the problem of manually drawing and updating complex system diagrams.

How it works

The tool acts as an agent skill that integrates with LLM-powered coding assistants (like Claude Code or Cursor). It uses a multi-step pipeline: it interprets your request or analyzes source code (like Terraform, Kubernetes configs, or Python classes), plans the layout, generates the .drawio XML, and performs self-correction on the visual output. It also includes a suite of CLI tools for importing existing infrastructure, enforcing architecture rules in CI, and syncing changes to diagrams without destroying manual layout adjustments.

Who it’s for

It is designed for software engineers, architects, and DevOps professionals who want to document their systems, visualize code structures, or manage infrastructure diagrams as code without the overhead of manual drawing tools.

Highlights

  • Automated Generation: Converts natural language prompts and code sources (Python, Go, Rust, Terraform, K8s) into structured, editable diagrams.
  • Incremental Sync: Updates diagrams based on code changes while preserving your manual layout and styling tweaks.
  • Architecture-as-Code: Enforces architecture rules (e.g., dependency cycles, trust boundaries) in CI/CD pipelines with visual diffing.
  • Extensive Ecosystem: Supports 11+ diagram types (C4, UML, ERD, etc.), includes 10,000+ official icons, and provides 321 AI/LLM brand logos for modern tech stacks.
  • Agent Compatibility: Functions as an MCP server or standalone CLI, making it compatible with major AI coding agents like Claude Desktop and Cursor.

Related

  • Project
    DayuanJiang/next-ai-draw-io

    An AI-powered diagramming tool that uses LLMs to create and modify draw.io diagrams via natural language, image uploads, or document ingestion.

  • Project
    sparklabx/drawio-ai-kit

    drawio‑ai‑kit is a self‑contained CLI and skill that lets LLM agents generate validated, standards‑compliant draw.io cloud‑architecture diagrams. It ships a catalog of real stencil IDs, design‑principle rules, and a static validator to catch hallucinated icons. Installable with a one‑liner (Node ≥ 20.6 or Bun), it offers a TypeScript API (`Diagram`, `group`, `icon`, `renderTree`) for declarative layout, automatic coordinate computation, and optional PNG rendering. The kit covers AWS, Azure, GCP, Databricks, BPMN and hybrid scenarios, making AI‑driven diagram creation reliable and visually consistent.

  • Project
    openflowkit/openflowkit

    An open-source architecture diagramming tool that allows users and AI agents to create, edit, and synchronize diagrams with codebases via a shared DSL.

  • Project
    liujuntao123/smart-excalidraw-next

    An AI-powered diagramming tool that converts natural language descriptions into professional, editable Excalidraw charts using LLMs and a smart connection algorithm.

  • Project
    cathrynlavery/diagram-design

    A set of editorial-quality diagram templates and an AI agent skill that allows users to generate brand-consistent technical visuals using natural language.