AI is changing how people work with data. Not in the distant future. Right now. Large language models can summarize reports, write code, spot patterns in spreadsheets, and answer questions about complex datasets. But they have a fundamental limitation: they can only work with what they can see. And most organizational data sits behind logins, inside platforms, and across systems that AI cannot reach. This gap between what AI can do and what it can access is the bottleneck. Today, we are releasing the OpenHEXA MCP server to close that gap.
The Model Context Protocol (MCP) is an open standard, originally developed by Anthropic and now adopted by Google, OpenAI, and others. It solves a simple problem: how do you let AI assistants interact with external tools and data in a structured, secure way?
Think of MCP as a universal adapter. Before it existed, every AI tool needed a custom integration for every data source. MCP replaces that tangle of one-off connections with a single protocol that any AI assistant can use to talk to any compatible service.
For OpenHEXA users, this means your AI assistant can now work directly with your workspaces, datasets, pipelines, and files inside the platform. No copy-pasting. No manual exports. No context lost in translation.
The first version of the OpenHEXA MCP server exposes eight tools that let AI assistants:
This is the foundation. More tools will follow as we expand what the server can do.
The OpenHEXA MCP server uses OAuth authentication. When you connect your AI assistant, it opens a browser window where you log in with your existing OpenHEXA credentials. The server then operates with your permissions, and only your permissions. If you cannot access a workspace through the OpenHEXA interface, your AI assistant cannot access it either.
This matters. The AI sees exactly what you see. Nothing more. Your organization’s access controls, role-based permissions, and data governance policies remain fully in effect. AI becomes a new way to interact with your data, not a way to bypass the rules that protect it.
The MCP server works with major AI assistants including Claude, ChatGPT, and Gemini. Setup takes a few minutes.
Add the OpenHEXA MCP server to your AI assistant using your instance URL. Authenticate through the browser-based OAuth flow. Start asking your AI assistant about your OpenHEXA data. That’s it. Once connected, you can ask things like “List my OpenHEXA workspaces” or “What datasets are available in the malaria-surveillance workspace?” and get answers drawn directly from your platform.
Detailed setup instructions for each supported AI assistant are available at your OpenHEXA instance’s MCP page.
This release is version 0.0.1. It covers the essentials: browsing, reading, and writing. We are already working on expanding the server’s capabilities to cover more of what OpenHEXA offers, from creating pipeline runs to deeper dataset exploration.
We built OpenHEXA to bring data together for public health. The MCP server extends that mission by making your data accessible to the AI tools that are becoming part of everyday work. Securely, on your terms, within the permissions you have already set.
For more information: Contact us
