🛡️ PhishNet AI

A full-stack AI-powered URL detection ecosystem that identifies phishing, malware, and defacement attacks in real-time using machine learning and heuristic analysis.

🏆 Achievements

• 🥇 Winner: Project Exhibition 2025-26, Dept of AI & DS, KSSEM.
• 🥉 3rd Place: IEEE National Level Project Exhibition at CMRIT.

---

About The Project

PhishNet AI addresses the growing threat of zero-day phishing attacks where traditional blacklist-based defenses fail. This application aggregates threat intelligence using a hybrid engine that combines Machine Learning probabilities with heuristic checks.

The core strength of PhishNet AI lies in its browser extension, which acts as a proactive shield. Unlike passive scanners, the extension analyzes every tab you visit in real-time, ensuring users are protected before they interact with malicious content. All data is synced locally, guaranteeing privacy and persistence across browsing sessions.

Key Features

• ✅ Real-Time ML Prediction: Instantly classifies URLs as Benign, Phishing, Malware, or Defacement.
• ✅ Hybrid Analysis Engine: Combines Machine Learning probabilities with heuristic checks (IP detection, suspicious TLDs).
• ✅ Proactive Browser Extension: A powerful Chrome extension that automatically scans active tabs in the background, offering zero-click protection without disrupting workflow.
• ✅ Visual Analytics Dashboard: Interactive charts and probability breakdowns to understand threat levels.
• ✅ Local Storage Persistence: Scan history is automatically synced to local storage without server overhead.
• ✅ Full-Stack Architecture: Decoupled architecture with a Python backend and a React frontend.
• ✅ Responsive Design: Works seamlessly across desktop and mobile browsers.

Tech Stack

This project was built using a modern full-stack architecture within a monorepo.

• Frontend:

  • React (with Vite)
  • Tailwind CSS
  • Recharts for Data Visualization
  • Axios for API Calls

• Backend:

  • Python
  • FastAPI
  • Scikit-Learn for Model Inference
  • Uvicorn for Server
  • pickle for Model Serialization

• APIs & Tools:

  • Chrome Extensions API (Manifest V3)
  • Local Storage API

• Deployment:

  • Frontend: Vercel
  • Backend: Render
  • CI/CD pipeline managed from a single GitHub monorepo.

Getting Started (Local Setup)

To get a local copy up and running, follow these simple steps.

Prerequisites

• Node.js (v18 or later)
• Python (v3.8 or later)
• Git

Installation

1. Clone the repository:

   git clone [https://github.com/DarshanKumarA/PhishNet-AI-main.git](https://github.com/DarshanKumarA/PhishNet-AI-main.git)
   cd PhishNet-AI-main

2. Setup the Backend (/backend):

   • Navigate to the backend directory:

     cd backend

   • Install Python dependencies:

     pip install -r requirements.txt

   • Start the server:

     uvicorn main:app --reload

3. Setup the Frontend (/frontend):

   • From the root directory, navigate to the frontend directory:

     cd frontend

   • Install NPM packages:

     npm install

   • Start the client:

     npm run dev

4. Setup the Extension (/extension):

   • Open Chrome and navigate to chrome://extensions.
   • Enable Developer Mode.
   • Click Load Unpacked and select the /extension folder.

🔑 Environment Variables

You will need to configure environment variables for the frontend to run.

1. Frontend (/frontend/.env):

VITE_API_URL=[http://127.0.0.1:8000](http://127.0.0.1:8000)