Leverages OpenAI's API to power the educhain library functionality for generating educational content.
EduChain MCP Integration Server
This project implements an MCP-compatible Flask server that integrates with the educhain Python library to dynamically generate educational content for Claude Desktop.
It provides three core educational tools:
🧠 Multiple-Choice Questions (MCQs)
📘 Lesson Plans
🃏 Flashcards (Bonus)
🚀 Features
📡 Exposes REST API endpoints for Claude Desktop integration
🔐 Uses
.envfile to securely manage OpenAI API keys📦 Lightweight and easy to deploy
🧪 Includes sample responses and test script
Related MCP server: Canvas MCP Server
🛠️ Technologies Used
Python 3.10+
Flask
EduChain (via
educhainpackage)dotenv (
python-dotenvfor secure API handling)Claude Desktop MCP Protocol (local JSON config)
📦 Setup Instructions
1. Clone the Repository
git clone https://github.com/abanindra3/educhain-mcp.git
cd educhain-mcp
2. Create and Activate Virtual Environment (Recommended)
python -m venv venv
source venv/bin/activate # On Windows: venv\Scripts\activate
3. Install Dependencies
pip install -r requirements.txt
pip install flask python-dotenv
4. Create .env File
Create a .env file in the root directory and add your OpenAI API key:
OPENAI_API_KEY=your_openai_key_here
5. Run the Server
python mcp_server.py
Server will run at:
📍 http://localhost:5000
6. Test the Endpoints
You can run:
python test_server.py
Or test manually using Postman/Hoppscotch with these POST endpoints:
Endpoint Purpose Sample Payload
/generate_mcqs Generate MCQs { "topic": "Python loops", "count": 5 }
/generate_lesson_plan Create a lesson plan { "subject": "Algebra" }
/generate_flashcards Generate flashcards (bonus) { "topic": "OOP in Java", "count": 6 }