🎓 BrightPeak Academy MCP Server

A FastMCP-based academic management system that demonstrates how a client and server communicate using the Model Context Protocol (MCP).

The project simulates a university management platform where the client interacts with the server to retrieve student information, manage grades, generate reports, and request AI-powered academic evaluations using MCP Sampling.

---

🚀 Features

• 📚 List all available courses
• 👨‍🎓 Retrieve student profiles
• 📝 Enroll students in courses
• ✏️ Update student grades with role-based authorization
• 📊 Generate academic reports with progress notifications
• 🤖 AI-powered student evaluation using MCP Sampling

---

🛠 Technologies

• Python 3.13
• FastMCP
• SQLite
• asyncio

---

📁 Project Structure

brightpeak-mcp-server/
│
├── Mcp-Server/
│   ├── server.py
│   ├── database.db
│   └── ...
│
├── client/
│   ├── client.py
│   └── test.py
│
├── requirements.txt
├── README.md
└── .gitignore

---

⚙️ Installation

Clone the repository

git clone https://github.com/your-username/brightpeak-mcp-server.git

Navigate into the project

cd brightpeak-mcp-server

Create a virtual environment

python -m venv .venv

Activate it

Windows

.venv\Scripts\activate

Install dependencies

pip install -r requirements.txt

---

Run the Project

Run the client

python client/client.py

The client automatically starts the MCP server using PythonStdioTransport.

---

Available MCP Tools

Tool	Description
list_all_courses	Returns all available courses
get_student_profile	Retrieves a student's profile
enroll_student	Enrolls a student in a course
update_student_grade	Updates grades with authorization
generate_academic_report	Generates an academic report with progress tracking
request_student_evaluation	Requests an AI evaluation through MCP Sampling

---

Progress Notifications

The server sends progress updates while generating reports.

Example:

Progress: 0%
Collecting student records...

Progress: 30%
Analyzing grades...

Progress: 70%
Generating final report...

Progress: 100%
Done

---

MCP Sampling

The project demonstrates MCP Sampling.

Instead of generating an academic evaluation inside the server, the server sends a prompt to the client.

The client (AI) generates the response and sends it back to the server.

Flow:

Client
    │
    │ call_tool()
    ▼
Server
    │
    │ ctx.sample(...)
    ▼
Client AI
    │
    │ Generate Evaluation
    ▼
Server
    │
    ▼
Client

---

Authorization

Grade updates require one of the following roles:

• INSTRUCTOR
• ADMIN

Unauthorized requests are rejected.

---

Example Output

========== Calling generate_academic_report ==========

Progress: 0% Collecting student records...

Progress: 30% Analyzing grades...

Progress: 70% Generating final report...

Progress: 100%

Academic report generated successfully.