
next-ai-draw-io
AI assistant that draws and edits draw.io diagrams from chat, for engineers and designers
The repository was last updated yesterday and is published under Apache-2.0.
- Pricing
- Open source
- Platforms
- Web, macOS, Windows
- GitHub stars
- 37K
- Last commit
- 1 day ago
- Language
- TypeScript
- License
- Apache-2.0
Similar MCP servers
About next-ai-draw-io
next-ai-draw-io is an open-source MCP server for ai models and llm tooling. It is available on Web, macOS, Windows, Linux, API, CLI, vscode and self-hosted.
Rather than starting from an empty canvas, you describe the diagram you want and the AI places shapes and connectors on a real draw.io canvas. The result remains an ordinary draw.io file, so you can keep editing it by hand, rename elements, add pages and export to .drawio, .png or .svg.
Each AI change is saved as a version with a thumbnail, so you can compare it with the current canvas, restore it or undo it. Selecting a few shapes lets you ask for changes limited to that selection. The README also describes uploading a screenshot, sketch, PDF or text file for the AI to turn into a diagram.
The project ships as a web app, a desktop installer for Windows, macOS and Linux, and an MCP server that lets agents such as Claude Code, Cursor and VS Code draw diagrams directly. It can be self-hosted, and the README lists 25 model providers, including Ollama for local models.
What it does
- Chat-driven diagram generation
- Describe a flowchart, architecture or sequence diagram in a sentence and the AI draws it on the canvas.
- Selection-scoped edits
- Select shapes on the canvas and ask the AI to change only those elements.
- Version history and undo
- Every AI change is stored as a version that can be compared, restored or reverted.
- Cloud icon libraries
- Built-in AWS, Azure, GCP and Kubernetes icon sets are available for technical diagrams.
- Multi-provider model support
- Works with 25 listed model providers, and you can supply your own API key in the browser.
- Self-hosted deployment options
- The README documents local, Docker, Vercel, Cloudflare and offline deployment routes.
Good for
- Drafting AWS or Kubernetes architecture diagrams from a written description
- Turning a whiteboard photo or sketch into an editable draw.io diagram
- Producing BPMN process or ER diagrams for documentation
- Letting an AI coding agent generate diagrams inside its workflow via MCP
Pricing
Open source
Free and Apache-2.0 licensed; the online demo has a usage limit that can be lifted with your own API key.
Prices change. Check next-ai-drawio.jiang.jp before you buy.
next-ai-draw-io FAQ
Is there a usage limit on the online demo?
Yes, the README says the demo has a usage limit, which can be lifted by entering your own provider and API key in the settings.
Where is the API key stored?
The README says the key stays in your browser and is never sent to the server for storage.
Which license does the project use?
The repository is licensed under Apache-2.0.
How do I add it to Claude Code?
The README gives the command claude mcp add --scope user drawio -- npx @next-ai-drawio/mcp-server@latest.
Is next-ai-draw-io free?
Yes. next-ai-draw-io is open source, so you can use and self-host it for free.
What are the best alternatives to next-ai-draw-io?
The closest alternatives to next-ai-draw-io are draw.io, Diagrams.so, AI Flowchart, FlowchartAI and LetDraw.
Where this comes from
- Description: written from the project's README and the project's website.
