Hi all, I’d like to share a pet project I have been working on over the past few months, and I believe it may be of interest to the OFBiz community.
Model Context Protocol (MCP) [*] is an emerging open protocol designed to allow AI-driven applications to interact with external tools and data sources using natural language prompts. Its goal is to standardize how Large Language Models (LLMs) can call external APIs or services in a structured and secure way. I have been exploring how MCP could be applied to Apache OFBiz, and as part of this effort I created a GitHub project implementing an MCP server for OFBiz: https://github.com/jacopoc/mcp-server-for-apache-ofbiz The MCP server communicates with OFBiz through RESTful API calls and is intended as a lightweight platform on which new MCP tools can easily be implemented. Out-of-the-box it currently includes a template tool that calls OFBiz’s "findProductById" service, but the architecture is designed to make the addition of new tools straightforward. The idea is that AI applications such as Claude, ChatGPT, Gemini, or custom AI agents can "reason" and leverage these tools to interact with OFBiz, for example: “Create a SEO-friendly description of the product WG-1111.” Some additional project highlights: * The MCP server is intentionally decoupled from OFBiz and can be deployed either locally or as an external service. * It works with ootb OFBiz, no patches required. * The project has reached a first milestone: it is now fully compliant with the latest MCP specification, and I’ll keep extending it with more features and tools. I am happy to continue maintaining it as an independent project, but I am open to discussing a home for it within the OFBiz community if there is interest. If you’re curious, please take a look, I would really appreciate feedback, ideas, or contributions. I can also provide the URL of a running instance that you can easily attach to an AI application (e.g., Claude) for testing. Just let me know if you’d like access or if the README isn’t sufficient to get you started. Thanks for your time, and looking forward to your thoughts! Best regards, Jacopo [*] https://modelcontextprotocol.io/docs/getting-started/intro
