How MCP transforms what’s possible with AI in Claris FileMaker
The introduction of Claris MCP in December 2025 opened the door to a host of new possibilities for AI and custom software. Just like Neo in “The Matrix,” who can learn new skills and abilities in moments, MCP allows your LLMs access to additional capabilities that weren’t possible before.
Here’s a brief overview of what Claris MCP adds to databases and custom applications built on the Claris Platform.
What is Model Context Protocol (MCP)?
MCP (Model Context Protocol) is an open protocol introduced by Anthropic designed to standardize the ways applications provide context to large language models (LLMs) via secure, structured communication between data sources and AI clients. Operating on a three-component model (host, client, and server), this architecture allows AI clients to access real-time data securely, then perform actions across different platforms without requiring a custom integration for each service.
Claris MCP is an MCP server designed to connect FileMaker databases with AI assistants and other hosts that are compatible with this protocol. Instead of needing to build a custom integration for every AI tool you want to connect to your data, Claris MCP offers a secure, standardized way for LLMs to understand and interact with your FileMaker data.
With MCP, you can build connections, select tables and scripts, configure database tools, and generate configuration snippets – creating seamless integrations and allowing you to deploy AI quickly and effectively.
What Claris MCP makes possible
On a recent Claris Community Live, Michael Wallace and Yichen Yang presented several detailed, practical use cases for Claris MCP and FileMaker, including:
- Using Anthropic’s Claude to generate detailed, realistic test data for a CRM database, including contacts and sales records
- Again using Claude, query your FileMaker database to summarize information and gather data using natural-language prompts
- Analyze data on team utilization to generate dynamic dashboards, ensuring jobs are assigned fairly and that workload doesn’t impact service delivery
- Perform an in-depth analysis of your financial data, identifying patterns in costs and revenue and delivering strategic insights
- Create new work orders via a FileMaker script – for example, when a new service call comes in, automatically create a new job order and assign it to the technician who has the greatest availability.
For developers as well as end users, there’s good inspiration to be found in this session. But if you already know there’s a challenge that AI could help address, get in touch today and we’ll ideate to help you understand which tool is the best fit for your needs.
Getting started with Claris MCP
The basic setup of Claris MCP can be done in just a few minutes and requires only a small amount of configuration and no coding to get started.
You need a Claris ID account that gives you a manager role in Claris Studio, and you need to be running FileMaker Server 2025 (version 22). In addition, you’ll need to have OData and FileMaker Data API enabled in your Server Admin Console. The account you’re using also needs to have fmrest and fmodata extended privileges enabled as well. Finally, FileMaker file accounts need access to the connected tables and fields that you want to make available to the MCP client.
Once that’s done, you can create a context, add a connection to your FileMaker database, and generate an MCP configuration snippet – that is, a JSON object with a URL and a static token – for connecting to your AI client.
If you have additional questions or want to explore a potential use case at your organization, reach out to our team. We’ll not only help you learn more about Claris MCP, but ask questions to better understand how LLMs, custom software, and other tools could help you solve solve problems and work more efficiently.
