How Can AI Better Understand VEX? I built an MCP Server for VEX Robotics

VEX V5 Robotics Competition
1

What is MCP Server?

MCP (Model Context Protocol) is an open standard that connects AI applications with external systems. Through MCP, AI tools like Claude or ChatGPT can link to data sources (local files, databases), tools (search engines, calculators), or workflows (prompt templates).

Think of MCP as the USB-C port for AI applications: just as USB-C provides a universal connection standard for devices, MCP provides a standardized way for AI apps to connect with external systems.

Current Industry Usage

MCP already has practical use cases in industry. For example, Anthropic enabled MCP in Claude, allowing developers to connect Claude with tools like Google Drive, Slack, and GitHub for knowledge retrieval and team collaboration.

Building on this foundation, we created a VEX MCP Server. The goal is to help AI understand robotics competitions more deeply.

Currently, the server supports retrieving competition and team information from RobotEvents, which means AI agents (like Claude or ChatGPT) can naturally integrate event data into conversations and provide accurate answers.

Demo Capabilities

  • :magnifying_glass_tilted_left: Search Teams
  • :date: Get Event Info & Results

Watch the demos here:

With the VEX MCP Server, AI agents can fetch official RobotEvents data in real time and present it in natural language — making the process of finding team and competition info much easier.

Open Source Project

This project is open-sourced on GitHub and also published to npm.
:backhand_index_pointing_right: GitHub Repository: yangjack8192/vex-mcp-server

The repo includes detailed setup instructions and examples, so you can quickly integrate it into any MCP-enabled agent (Claude, GitHub Copilot, etc.).

Roadmap & Improvements

There’s still room for improvement, including:

  • Enhancing search accuracy
  • Adding Skills Challenge ranking lookups
  • Integrating more resources beyond RobotEvents

Join the VEX Community Effort :tada:

We’d love for the VEX community to join this project! You can contribute in several ways:

  • :light_bulb: Share feature requests: What do you wish this MCP Server could answer? Post in the comments!
  • :lady_beetle: Report issues: If you encounter any problems, reach out via email [redacted] or open a GitHub issue.
  • :hammer_and_wrench: Contribute code: Submit PRs directly on GitHub, or contact me via email to collaborate.

Let’s build a professional, useful, and extensible AI-powered VEX robotics tool library together :rocket:

2

moderation removed an email address from this post.

10 Likes
3

@8192D-YANG

That’s a good initiative.

It has been in my mind for a while on using AI for my teams. But I was thinking more of using RAG (to serve as a knowledge base for students to do their own self-directed learning).

But I do think it might be a good idea to get MCP and RAG to work together.
And it will also be able to extend your current usage of just searching for teams’ info and competitions’ data.

4

I wonder how many boxes you can check off just from this one post…

5

I’ll just leave this here…

IMG_3603
IMG_36031179×872 84.1 KB

6

somewhat agree, except the OP has (apparently) been on the forum for ~12 years.

7

And also, 8192 is from China.
Maybe they used AI to help with the translation from chinese to english?

@9MotorGang and @lewis care to share what are your concerns about this post?

For me, I was just focusing on the content of the post and the MCP server that is being setup.

8

The primary concern is just that it sets off all the alarm bells for AI generated content

9

Generally, LLM is good at phrasing and formatting the languages, the purpose of using LLM in the main post is to help present the context in a most understandable way. Let me know if you have different ideas or concerns.
( I do not to deviate this post to much, I feel like the AI discussion can probably have its own post)

10

Yeah I agree that would be a great idea. We could definitely add one tool so that the agent defines keywords and use it to helps people look up for documentation through RAG. Is this a ongoing project? Do you have a timeline for your RAG project? I think there is definitely room for collaboration.

11

This is a really cool project, because having an MCP server for VEX makes it much easier for AI tools to pull official team and event data directly from RobotEvents instead of users searching manually. Real-time team lookup and event results are genuinely useful, and the open-source approach means the community can help improve features, accuracy, and future additions like Skills rankings and more integrations.