🎓 Vishwalata College AI Assistant

A highly responsive, real-time AI-powered admission counselor and chat assistant built exclusively for Vishwalata College. This intelligent web application guides students, provides course details, showcases campus facilities, and seamlessly generates admission inquiry leads.

🌟 Key Features

• ⚡ Real-time Streaming AI Response: Powered by WebSockets (Flask-SocketIO), ensuring users see the AI typing its responses instantly.
• 🤖 Multi-LLM Support Engine: Easily switch between top-tier AI models including Google Gemini, Groq, and Sarvam AI directly via the backend configuration.
• 🏫 Intelligent Data Contextualization: Automatically fetches information from a robust SQLite database carrying accurate details regarding college fees, facilities, and placements.
• 🌐 Multi-Lingual Context: Automatically tailors the AI personality to switch intuitively between English, Marathi, and Hindi, accommodating local students seamlessly.
• 📸 Dynamic Visual Gallery Integration: Auto-matches user queries regarding "campus", "library", or "hostels" and effortlessly displays corresponding visuals within the chat UI.
• 🔐 Bulletproof API Security: 100% Secure implementation. Employs strictly HttpOnly session cookies for state authorization—hiding internal logic completely from client vectors.
• 👨‍💻 Comprehensive Admin Panel: A secure interface strictly available to authorized admins to manually update/add course fees, new gallery images, and view actively collected inbound student inquiry leads.

🚀 Tech Stack

• Backend Developer: Python Flask
• Real-time Communication: Flask-SocketIO & Eventlet
• Database Management: Serverless SQLite3
• Security & Encryption: bcrypt, HttpOnly Session State
• Frontend UI Engine: Custom HTML5, Responsive Vanilla CSS3, JS Vanilla (No Framework Overhead)
• AI Integration: google.generativeai, groq, sarvamai

🛠️ Quick Installation Guide

Prerequisites

1. Python 3.9+ installed on your system.
2. An active API key from the supported LLM providers (Gemini / Groq / Sarvam).

1. Clone the repository

bash
git clone https://github.com/ganesha-raut/vishwalata_collage_Ai.git
cd vishwalata_collage_Ai

2. Setup the Virtual Environment

bash
python -m venv .venv
# On Windows
.venv\Scripts\activate
# On macOS/Linux
source .venv/bin/activate

3. Install Dependencies

bash
pip install -r requirements.txt

4. Setup API Keys

Open ai_models.py and input your respective API Key.

python
ACTIVE_MODEL = "sarvam" # or "gemini", "groq"

5. Run the Server

bash
python app.py

The application will safely initialize and serve the app synchronously via WebSockets on http://127.0.0.1:5000.

🔒 Security Posture

• All database interactions utilize parameterized executions, neutralizing SQL-Injection.
• Frontend completely abstracted away from third-party Application keys.
• Passed comprehensive Black-Box UI Analysis. (See SECURITY_TEST_REPORT.md for full breakdown).

Developed & Maintained by Ganesha Raut.