lgazo/drawio-mcp-server

What it solves

This project provides a way for AI agents to programmatically create, modify, and manage diagrams in Draw.io (Diagrams.net). It eliminates the need for users to manually draw architectural diagrams, flowcharts, and visual documentation by allowing an AI assistant to handle the layout and element placement.

How it works

It operates as a Model Context Protocol (MCP) server. It can be used in two primary ways:

1. Built-in Editor: The server hosts a local instance of the Draw.io editor directly in the browser.
2. Browser Extension: It connects to an existing Draw.io session running in a browser via a dedicated extension.

Once connected, the AI agent uses a set of MCP tools to perform actions such as adding shapes, managing layers, creating pages, and routing edges between elements.

Who it’s for

Developers, architects, and technical writers who use MCP-compatible AI clients (like Claude Desktop, Claude Code, or Zed) and want to AI-generated visual documentation.

Highlights

• Comprehensive Diagramming Tools: Tools for document discovery, shape/edge modification, page management, and layer organization.
• Vendor Stencil Support: Automatically discovers stencils for AWS, GCP, Azure, and Cisco at runtime.
• Multi-document/Page Workflow: Supports targeting specific documents and pages for complex, multi-tab workflows.
• Mermaid Integration: Ability to import, embed, or expand Mermaid diagrams.
• Flexible Deployment: Includes a built-in editor, a browser extension, and support for TLS/HTTPS for secure connections.
• File Export: Supports importing and exporting diagrams as XML, SVG, or PNG.