"How much unfunded capital do we have across Fund III?"
Until now, getting that answer meant running a report, exporting a spreadsheet, or tracking down the one person who knew where the number lived. It was a manual process that slowed decisions and pulled teams away from higher-value work.
That's changed. Entrilia now connects directly to MCP-compatible AI assistants through Entrilia MCP. Whether your team uses ChatGPT, Claude, Glean, or another MCP-compatible platform, you can ask questions in plain language and get answers grounded in your live fund accounting data.
The Challenge: Probability vs. Precision
Large language models (LLMs) are very good at reasoning through complex problems. What they can't do on their own is know your data.
Without a direct connection to your fund accounting system, an AI has no access to your actual figures. It works from general knowledge instead. In fund accounting, that's not good enough. A number that's probably right is not a number you can send to a GP.
That's the problem Entrilia MCP solves. It gives the AI direct, governed access to your live Entrilia data, so every answer is grounded in your books, not a best guess.
The Foundation: Where the Numbers Come From
Every answer the MCP Connector returns traces back to SmartViews, Entrilia's reporting and analytics layer. SmartViews is what makes the MCP Connector different from a generic data integration.
Most AI integrations expose static data, pre-run reports or fixed exports that reflect a point in time. They give the AI numbers, but no understanding of what those numbers mean or how they were calculated. The AI has to interpret them without context, which introduces risk.
SmartViews works differently. It is a context layer built specifically for fund operations, a verified model of your fund data that already understands the logic behind every metric. When the MCP server receives a question, it queries SmartViews directly. If the right report already exists, it pulls from it. If it doesn't, SmartViews generates one on demand to answer your specific question.
Either way, the AI is not guessing. It works from figures that Entrilia has already defined and verified, and every answer comes with a direct link to the source report. Nothing is a black box.
The Protocol: A Familiar Idea, Upgraded
The Model Context Protocol, or MCP, is an open standard released by Anthropic for connecting AI agents to software. Think of the Entrilia MCP server as a UI for AI agents. The same way you come to Entrilia through a browser and interact with your fund data, an AI agent does the same thing through the MCP server. It can ask questions, retrieve reports, and as the product evolves, trigger actions directly inside Entrilia.
Because Entrilia supports the protocol itself rather than a single AI tool, the connector works with any MCP-compatible platform , such as Claude, ChatGPT, Glean, or your own internal orchestrator. One integration, any AI platform.
The Trust Model: Transparency and Verification
Getting a fast answer from an AI is straightforward. Knowing whether that answer is right is harder. That is where most AI integrations fall short.
Every response from the Entrilia MCP server includes a direct link to the report behind the answer. Open it, and you land in SmartViews, where you can see the data visually and verify every figure the AI returned. You do not have to take the AI's word for it. The source is one click away.
Today, our MCP server is read-only. It retrieves and returns data, without modifying anything in Entrilia. Edit mode is already in development, more on that below.
The Controls: Your Data, Your Rules
Fund environments have specific requirements around data accuracy, access control, and privacy. Entrilia MCP is built with those requirements in mind.
Your data stays in Entrilia. The MCP server does not duplicate, migrate, or store your fund data anywhere else. It queries Entrilia live and returns only the figures needed to answer each question.
Your prompts are never used for training. Neither your questions nor the data returned are used to train any third-party AI model, under the enterprise data protections of the AI provider you have chosen.
You control what AI can access. Each user can only query data they already have permission to view in Entrilia. Administrators can go further, applying additional restrictions specifically for AI access, including controls over what sensitive data, such as investor names, contact details, or bank information, the AI assistant can see.
This Is Just The Beginning
The way teams use AI is evolving every week, from ad hoc queries to automated reconciliation, custom dashboards, and scenario modelling. What we're building is the context layer that makes all of it possible: a single, verified connection between your fund data and any AI tool your organization chooses to use.
The current version of Entrilia MCP is focused on retrieval, giving your team instant, verifiable answers from live fund data. But that is just the beginning of what the Entrilia context layer makes possible.
The next step is allowing users to make changes directly from their AI assistant—updating inputs, adjusting figures, and acting on the data without switching back to the platform. Ask a question, spot something that needs updating, and change it in the same conversation.
If you're exploring what a modern fund accounting platform can do for your organization, we'd love to show you. Book a demo to see it in action.
