Introducing the Ellie MCP Server: Bridging AI and Data Modeling

Release

Product

Abstract:

The Ellie MCP Server is a new integration that utilizes the Model Context Protocol (MCP) to enable AI assistants (such as Cursor and Windsurf) to interact directly with your actual data models in Ellie. This bridges the gap between large language models and the data modeling workflow, transforming AI from a disconnected chatbot into a governed "modeling partner." The server acts as a secure gateway, allowing AI to perform real operations—like reading existing architectures, programmatically creating/updating models, and refining details—all while enforcing Ellie's existing governance, permissions, and versioning rules via the underlying Ellie API. This results in rapid prototyping, reduced context switching, and the creation of governed, versioned data assets directly within your Ellie repository.

We are proud to announce the release of the Ellie MCP Server, a new integration that bridges the gap between LLMs and your actual data models in Ellie.

Until now, using AI for data modeling often meant a disconnected workflow: chatting with a bot in one window, and manually building models in another.

The Ellie MCP Server changes this by providing AI assistants—such as Cursor, Windsurf, or Claude Desktop—with the ability to interact directly with Ellie’s platform. By utilizing a standardized set of tools via the Model Context Protocol (MCP), it allows LLMs to perform real operations on your data models, not just talk about them.

How It Works

The MCP Server acts as a translator and a secure gateway. It gives your AI assistant a specific toolkit to:

Read and Understand: The AI can scan your existing models to understand your current architecture, ensuring new additions fit your standards.
Create and Update: It can programmatically generate new models, entities, and attributes based on your natural language prompts.
Refine Details: It can bulk-edit definitions, standardize attribute names, and manage metadata without manual data entry.

Example how Ellie MCP Server can help Data Engineers understand a model from any tool

Why this matters

By allowing LLMs to perform operations on models, entities, and attributes directly, we are moving from "AI as a chatbot" to "AI as a modeling partner."

Rapid Prototyping: Drastically reduce the time between requirements gathering and model design. With Ellie MCP, AI can interpret your specs and immediately populate your Ellie folders with detailed Conceptual, Logical, and Physical drafts.
Reduced Context Switching: You don't need to copy-paste JSON or text descriptions between tools. The AI works where you work—inside the model.
Standardized Operations: Because the MCP server uses a standardized set of tools, it ensures that the AI interacts with Ellie predictably and safely, respecting the structure of your repository.