dgm003/mcp_educhain_assignment

EduChain MCP Assignment

Overview

This project implements an MCP (Model Context Protocol) server using the EduChain library. The server exposes endpoints to generate educational content such as lesson plans and multiple-choice questions (MCQs), designed for integration with Claude Desktop or other MCP-compatible clients.

Features

• Generate comprehensive lesson plans for any topic.
• Generate multiple-choice questions (MCQs) for any topic.
• Easily extensible to add more educational tools/resources.
• Fully documented code with clear comments and citations.

Setup Instructions

1. Clone the Repository

git clone https://github.com/dgm003/mcp_educhain_assignment.git
cd mcp_educhain_assignment 

2. Install Dependencies

It is recommended to use a virtual environment:

python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate
pip install -r requirements.txt

3. Run the MCP Server

uvicorn main:app --reload

The server will start at http://127.0.0.1:8000 by default.

Usage

Testing the Endpoints

You can test the endpoints using Thunder Client (a VSCode extension for API testing), Postman, or any REST client of your choice.

To use Thunder Client:

• Install the Thunder Client extension in VSCode.
• Open the Thunder Client tab in VSCode.
• Create a new POST request to the desired endpoint (e.g., http://127.0.0.1:8000/generate_lesson_plan).
• Set the request body to JSON and provide the required fields (see examples below).
• Click "Send" to view the response.

Endpoints

• POST /generate_lesson_plan
  Generate a lesson plan for a given topic.
• POST /generate_mcqs
  Generate multiple-choice questions for a given topic.

Example Requests

Generate Lesson Plan

POST http://127.0.0.1:8000/generate_lesson_plan
Content-Type: application/json
{
  "topic": "Algebra",
  "duration": "60 minutes",
  "grade_level": "High School"
}

Generate MCQs

POST http://127.0.0.1:8000/generate_mcqs
Content-Type: application/json
{
  "topic": "Python loops",
  "num": 5
}

Sample Responses

See Sample_Responses.txt for example outputs from the endpoints.

Screenshots

Terminal - FastAPI Server Running

Thunder Client - Successful MCQ Response

Thunder Client - Successful Lesson Plan Response

Deliverables

• Python code with clear comments and documentation.
• Sample commands and responses (Sample_Responses.txt).
• This README file.

Claude Desktop MCP Integration Note

The public version of Claude Desktop (from claude.ai) does not currently support MCP server integration or custom tool configuration. As a result, direct integration and testing with Claude Desktop was not possible. All endpoints have been tested using Thunder Client (VSCode extension) and Postman. Please contact the assignment coordinator if an MCP-enabled build of Claude Desktop is required.

Citations & References

• EduChain library
• EduChain documentation
• Model Context Protocol (MCP) Quickstart
• FastAPI documentation
• LangChain documentation

All code is original except where noted, and external libraries are cited above.

Contact

For any questions or guidance, please contact the assignment coordinators:

• prathmesh@buildfastwithai.com
• shubham@buildfastwithai.com