College AI Assistant

An interactive assistant for managing and querying college student information (profiles, marks, attendance) using an MCP-based tool server and Google Gemini.

Overview

College AI Assistant demonstrates a small local system with:

• an MCP tool server (server.py) exposing student data and helper tools
• a CLI assistant (ai_agent.py) that uses Gemini plus MCP tools
• a FastAPI wrapper (api.py) for the browser UI
• a simple static frontend in web/
• a local SQLite database managed by database.py

Features

• Search student profiles by name
• Retrieve detailed student information by ID
• View subject-wise marks, average marks, and academic performance
• View subject-wise attendance and attendance analysis
• Find students with attendance below a threshold
• Run the assistant from CLI, HTTP API, or web UI

Repository Structure

• ai_agent.py — CLI client that starts an interactive assistant session
• api.py — FastAPI app exposing /chat and health endpoints
• server.py — MCP tool server with student data tools and prompts
• database.py — SQLite helpers, schema creation, and sample data seeding
• config/settings.py — Gemini and MCP server configuration
• agent/ — shared adapter code for integrating Gemini and MCP tools
• web/ — static frontend: index.html, script.js, style.css
• requirements.txt — Python dependencies

Requirements

• Python 3.9+
• A valid Gemini API key

Install dependencies with:

pip install -r requirements.txt

Configuration

Create a .env file in the project root with:

GEMINI_API_KEY=your_gemini_api_key_here

Default Gemini model and MCP settings are configured in config/settings.py:

• GEMINI_MODEL = "gemini-2.5-flash"
• MCP_SERVER_COMMAND = "python"
• MCP_SERVER_FILE = "server.py"

Setup & Run

1. Initialize the database and seed sample data:

python database.py

2. Start the MCP tool server:

python server.py

3a. Run the CLI assistant:

python ai_agent.py

3b. Or start the HTTP API for the web UI:

uvicorn api:app --reload --host 127.0.0.1 --port 8000

4. Open web/index.html in your browser. The UI sends chat requests to http://127.0.0.1:8000/chat by default.

API Endpoints

• GET / — basic health check
• GET /health — service status
• POST /chat — chat endpoint

Request body:

{ "message": "Hello", "session_id": "optional-session-id" }

Response body:

{ "session_id": "...", "response": "..." }

Notes

• The local SQLite database file is college.db.
• The web UI is a static front end and requires the FastAPI server to be running.
• database.py can be re-run to recreate tables and insert sample rows.

Troubleshooting

• If the web UI cannot connect, verify the API is running at 127.0.0.1:8000.
• If Gemini fails due to quota or authentication, check GEMINI_API_KEY and your Gemini account settings.

Optional Improvements

• Add docker-compose.yml for streamlined startup
• Add a Makefile or PowerShell script for common commands
• Add tests for the API and MCP tools