Hi Stephen, Apologies for the late reply. Thanks Stephen for getting the ball rolling here, and thanks Pritam for the contribution. I think it would be great to have an official MCP as part of the project, and Pritam’s work looks like a great start.
I see the PR is up already, I’ll try to get a review in soon. I don’t have any large concerns here. Once the PR is merged, it will be nice to create a “mcp” component in JIRA to begin tracking bugs and features for the project. Thanks Cole From: Stephen Mallette <[email protected]> Date: Thursday, August 28, 2025 at 10:06 AM To: [email protected] <[email protected]> Subject: [DISCUSS] gremlin-mcp I've had a word with Pritam Kadam who is in the relatively early stages of bringing Gremlin to the MCP world. Pritam has indicated interest in collaborating on that work here to form an official Apache TinkerPop supported gremlin-mcp module. I believe that this is one of the areas related to AI that the TinkerPop Community has an interest in and that Pritam's contribution would be a welcome addition. Basic features Pritam has been working on are: "What's the structure of my graph?" - Automatic schema discovery "Show me all users over 30 and their connections" - Complex graph queries "Find the shortest path between Alice and Bob" - Relationship analysis "Give me graph statistics and metrics" - Data insights "Import this GraphSON data" - Data loading "Export user data as CSV" - Data extraction Smart enum discovery - AI learns your data's valid values automatically Assuming this direction makes sense for everyone, Pritam could move forward with a pull request with the basics for their work. I'd suggest that this go to 3.8-dev and, for now, go under gremlin-tools as a sort of separate project until it can be integrated into the build in whole and be made ready for a proper release at some point in the future (which may not be 3.8.0). Happy to hear any thoughts or concerns. Warning: The sender of this message could not be validated and may not be the actual sender.
