UTD Nebula MCP Server

A Model Context Protocol (MCP) server that bridges LLMs with University of Texas at Dallas course data and RateMyProfessors ratings.

🚀 Overview

This project implements an MCP Server designed to empower UTD students and faculty by connecting Large Language Models (LLMs) like Claude or ChatGPT to real-time university data.

Finding the right class and the right professor typically involves juggling multiple tabs: CourseBook for schedules, Nebula for details, and RateMyProfessors for reviews. This tool unifies that experience. You can simply ask your AI assistant:

"Who is the highest-rated professor teaching Artificial Intelligence next semester?" "Find me an easy 3000-level elective in the CS department."

✨ Features

• Nebula API Integration: Fetches official UTD course information, including sections, course descriptions, schedules, and prerequisites, directly from Nebula Labs' API.
• RateMyProfessors (RMP) Scraping: Retrieves detailed professor metadata via GraphQL, including:
• Metrics: Average Rating (1-5), Difficulty Score, "Would Take Again" %.
• Reviews: Specific written student feedback.
• Tags: Attributes like "Tough Grader", "Inspirational", or "Lecture Heavy".

🛠️ Architecture

This server follows the Model Context Protocol (MCP) standard.

1. Client: The AI interface (e.g., Claude Desktop, Cursor, or a custom MCP client).
2. MCP Server (Go): This application. It exposes tools like get_course_sections and get_professor_reviews.
3. Data Sources:

• Nebula API: Source of truth for UTD academic data.
• RateMyProfessors: Source for social sentiment and ratings.

📦 Installation

Prerequisites

• Go (1.21 or higher)
• Access to UTD Nebula API (if an API key is required).

Build from Source

# Clone the repository
git clone https://github.com/lalipop05/utd-nebula-api-mcp-server.git
cd utd-nebula-api-mcp-server

# Install dependencies
go mod download

# Build the binary
go build -o utd-mcp-server ./cmd/server/main.go

Testing

You can test the MCP server using the MCP Inspector:

npx @modelcontextprotocol/inspector go run ./cmd/server/main.go

⚙️ Configuration

Create a .env file in the root directory or set environment variables directly. Set PATH_TO_ENV in main.go to be the absolute path to your .env file

# Example .env configuration
NEBULA_API_KEY=your_nebula_api_key_here

🔌 Integration (How to Use)

With Claude Desktop

For the most up to date information on how to connect this server to Claude Deskop, consult official MCP website

To use this with the Claude Desktop app, add the following to your claude_desktop_config.json:

{
  "mcpServers": {
    "utd-nebula": {
      "command": "/absolute/path/to/utd-nebula-api-mcp-server/utd-mcp-server.exe",
      "args": []
    }
  }
}

With Other MCP Clients

Run the server executable directly. It communicates over Stdio (Standard Input/Output) by default, fully compliant with the MCP specification.

💡 Example Prompts

Once connected, your LLM gains "knowledge" of UTD. You can ask:

• Course Search: "I am looking for a class in Embedded Systems. What are my options for next semester?"
• Professor Comparison: "Who is the best professor to take for CS 4365 (Artificial Intelligence)? Compare their difficulty and ratings."
• Deep Dive: "Does Professor Smith require a textbook? What do recent reviews say about their attendance policy?"
• Schedule Planning: "Find me a CS guided elective that fits on Tuesday/Thursday afternoons."

🧩 Project Structure

• cmd/: Entry point for the application.

• internal/nebula/: Client and logic for interacting with UTD Nebula API.

• internal/rmp/: Logic for RateMyProfessors GraphQL queries.

• data/: GraphQL templates and constants.

... View full README on GitHub