Make your AI coding assistant truly understand yFiles yFiles MCP Server
AI assistants are great — but without the right context they often generate outdated or incorrect code for yFiles. The yFiles mcp-server fixes this by giving your agent up-to-date, accurate knowledge about the yFiles for HTML SDK. It uses the open-source Model Context Protocol (MCP) to provide your AI agent efficient tools to search through the yFiles API, documentation and demo code.
With the MCP server the AI assistant has deep knowledge of yFiles concepts, patterns, and common development scenarios, making it an invaluable companion for both beginners and experienced developers.
Benefits of using the yFiles mcp-server
- Accurate and up-to-date information about the yFiles for HTML SDK
- Scaffolding new yFiles applications
- Code suggestions that follow best practices
- Easing the learning curve of the yFiles SDK
- Works with many popular AI agents via the Model Context Protocol (MCP)
Thank you!
Your message has been sent.
We'll get back to you as soon as possible.
Thank you!
Your message has been sent.
We'll get back to you as soon as possible.
Works with your favorite AI agent
We have successfully tested yfiles-mcp with the following AI agents. Thanks to the MCP standard, it should work with any agent that supports MCP servers.
In our experience, coding-focused LLMs like Claude Sonnet 4 work best for big scaffolding tasks where a full project setup is needed. GPT-5 has also been a solid choice for complex scenarios, while other models often shine when it comes to more specific, targeted prompts or questions.
Use cases
With the yFiles mcp-server, your AI assistant doesn't just explain how things work — it can directly implement features, scaffold projects, and answer API questions with real code. This lowers the learning curve for newcomers and speeds up everyday development for experienced users alike.
Getting started as a yFiles newcomer
yFiles for HTML provides outstanding documentation and more than 300 demos that illustrate almost every feature and scenario. These are invaluable learning and reference resources.
For newcomers, however, combining multiple concepts from the documentation into a working first project can feel overwhelming. The yFiles mcp-server lowers this barrier by letting your AI agent set up a functioning project directly, so you can see results right away and learn by example. Of course, this is typically not a single prompt, but rather a sequence of prompts where each one adds and combines more features.
Example prompt
Scaffolding new yFiles applications
When you're ready to start fresh or try out new ideas, the yFiles mcp-server can scaffold a project for you. It generates the boilerplate, initializes the GraphComponent, wires up interactions, and embeds common patterns—saving time on repetitive setup tasks so you can focus on your actual goals.
Example prompts for scaffolding
Accelerating development in your existing projects
Once your application is up and running, the real work begins: extending features, customizing styles, refining layouts, and improving performance. With its rich API, yFiles gives you full control over all of these aspects — but finding the right combination of classes, methods, and demos can sometimes be time-consuming.
The yFiles mcp-server, when paired with your AI agent, helps you move faster by implementing concrete code changes directly in your project context. Instead of spending time searching for the right demo or piecing things together manually, you can simply describe what you want to achieve, and your AI assistant generates the necessary code using the official yFiles APIs and best practices.
This dramatically speeds up iterative development, helps you avoid common mistakes, and shortens the time from idea to working feature.
Example prompts for existing projects
Answering yFiles API questions
The yFiles for HTML SDK is vast and powerful, offering hundreds of classes, interfaces, and specialized layout and visualization features. Even experienced developers can't (and shouldn't) memorize the entire API surface. Traditionally, when a question arises, you'd search through the documentation or look for the right demo — both excellent resources, but sometimes too slow when you're in the middle of coding.
With the yFiles mcp-server, your AI agent can directly answer API questions and implement code solutions immediately in your development environment. Instead of switching contexts to look up a class or an event, you simply ask: "How do I attach a listener for label clicks?" or "Which layout should I use for layered data?" — and your assistant can show you, explain it, and inject working code into your project.
This turns your AI agent into an interactive API companion: one that's always up-to-date, understands the entire yFiles SDK, and lowers the learning curve by translating abstract concepts into concrete, ready-to-use code.
How to get started
The yFiles mcp-server comes as part of the yFiles Dev Suite a development hub for working with yFiles. Aside from the MCP server, it helps you set up the SDK, explore the demos and documentation and also create stand-alone project from existing demos.
To make an AI agent aware of the yFiles mcp-server, you need to configure the MCP server in your AI agent's configuration. It can be started through the CLI:
npx --yes yfiles-dev-suite@latest --mcp
For Claude Code, the command would look like this:
claude mcp add yfiles-api -- npx --yes yfiles-dev-suite@latest --mcp
A common JSON configuration for the mcp server looks like this:
"yfiles-api": {
"command": "npx",
"args": [
"--yes",
"yfiles-dev-suite@latest",
"--mcp"
],
"transportType": "stdio"
}
System dependant
The exact command to start the yfiles-api mcp server depends on your operating system
and the tool you are configuring it for. For example, for some tools, on Windows you
may need to specify npx.cmd, while for most tools, as well as on macOS or
Linux it is sufficient to use npx.
This configuration enables your AI agent to communicate with the mcp tools for the yFiles API, providing access to specialized tools and knowledge for yFiles development.
Frequently Asked Questions
Which AI agents does the yFiles mcp-server work with?
Where can I get the yFiles mcp-server?
How much does the yFiles mcp-server cost?
How accurate and fast is the yFiles mcp-server when used with an AI agent?
The performance and quality of the yFiles mcp-server strongly depend on the AI agent and the underlying large language model (LLM) you are using. While the mcp-server always provides structured, up‑to‑date access to the yFiles SDK, the quality of the generated code and responses varies with the capabilities of the connected AI agent and model version.
How does the yFiles mcp-server help beginners?
How does the yFiles mcp-server help with scaffolding new projects?
How does the yFiles mcp-server support existing projects?
Recap
yFiles MCP server
The yFiles MCP server equips your AI coding assistant with up-to-date and reliable knowledge of the yFiles for HTML SDK, enabling it to deliver accurate code suggestions and use cases. By leveraging the Model Context Protocol (MCP) , it integrates seamlessly with popular AI agents to streamline tasks such as project scaffolding, feature implementation, API question answering, and many other use cases. The getting started guide explains how to connect and configure the MCP server with your AI agent for a smooth setup. These capabilities help both newcomers and experts efficiently build, customize, and maintain yFiles-based applications.
Related content
Got questions about
the yFiles MCP server?
Nicolas Brich
Senior Software Developer
Email: hello@yworks.com
Phone: +49 7071 9709050
11 reasons why web developers choose yFiles, the superior diagramming SDK
Choose your next steps
Get connected
Connect with our Customer Success Team regarding your ideas or projects.
Download yFiles
Try yFiles free of charge.
