NetShield

Your AI-Powered Shield Against Digital Threats

NetShield (formerly CodeGuardians) is a comprehensive cybersecurity command center designed to protect users from modern digital threats. It unifies advanced detection engines for identifying AI-generated deepfakes, phishing attacks, and malicious URLs into a single, intuitive platform.

🛡️ Core Features

1. 🔍 AI Image Detection

Goal: Detect AI-generated images, deepfakes, and synthetic media.

• Technology: EfficientNet-B0 Deep Learning Model (PyTorch)
• How it works:
  • Analyzes pixel-level artifacts and frequency patterns invisible to the human eye.
  • Distinguishes between real photos and those generated by GANs or Diffusion models (Midjourney, DALL-E, Stable Diffusion).
  • Provides a confidence score and probability map.

2. 🎣 Phishing Email Detector

Goal: Analyze emails and messages to detect social engineering and phishing attempts.

• Technology: Hybrid approach (ML + Rules + URL Analysis)
• Features:
  • ML Model: Text classification using TF-IDF and Logistic Regression trained on phishing datasets.
  • Rule-Based Engine: Detects urgency keywords ("verify", "suspend", "urgent") and suspicious patterns.
  • URL Scanner: Automatically extracts and scans all links within the message.

3. 🔗 URL Safety Scanner

Goal: Verify if a website link is safe before clicking.

• Technology: Hybrid Intelligent Service (Custom ML Model + Google Safe Browsing API)
• Features:
  • Machine Learning Analysis: Custom-trained model analyzing 19 structural URL features (domain entropy, suspicious TLDs, brand impersonation).
  • Real-time Verification: Cross-references links against Google's global blacklist.
  • Instant Verdict: Returns a combined "Safe" or "Unsafe" verdict with details.

4. ⚙️ Terminal Command Analysis

Goal: Protect developers and sysadmins from running dangerous shell commands.

• Technology: Static Analysis & Pattern Matching
• How it works:
  • Analyzes shell commands for destructive actions (e.g., rm -rf /, fork bombs).
  • Flags sudo usage, network connections, or obfuscated scripts.
  • Provides a safety explanation for each command part.

5. 🔐 Password Toolkit

Goal: Generate uncrackable passwords and check for breaches.

• Technology: Cryptographically Secure RNG + HaveIBeenPwned API
• Features:
  • Generator: Creates high-entropy passwords with custom length and character sets.
  • Strength Meter: Visualizes password complexity in real-time.
  • Breach Check: Hashes the password (k-Anonymity) and checks known data leaks.

6. 🧩 Browser Extension

Goal: Real-time protection while browsing.

• Technology: Chrome Extension V3 (JavaScript/HTML/CSS)
• Features:
  • Auto-Password Generation: Detects signup fields and suggests secure passwords.
  • Quick Phishing Scan: Highlight text or right-click to scan for phishing.

---

💻 Tech Stack

Backend

• Framework: Flask (Python)
• ML/AI: PyTorch, Scikit-learn, Pandas, NumPy
• Security: Google Safe Browsing API, HaveIBeenPwned API
• Server: Gunicorn

Frontend

• Design: Modern CSS3 (Glassmorphism, Dark Mode)
• Interactivity: Vanilla JavaScript
• Templates: Jinja2

Deployment

• Platform: Render / Vercel compatible
• Environment: Python Virtual Environment (.venv)

Installation & Setup

Prerequisites

• Python 3.8 or higher

1. Clone the Repository

git clone https://github.com/your-username/NetShield.git
cd NetShield

2. Create Virtual Environment

python -m venv venv

3. Activate Virtual Environment

# Windows (PowerShell)
.\venv\Scripts\Activate.ps1

# Windows (Command Prompt)
.\venv\Scripts\activate.bat

# Linux/macOS
source venv/bin/activate

4. Install Dependencies

pip install -r requirements.txt

5. Download Required Data

REQUIRED: Download the necessary models and datasets from the following link: Google Drive Data Link

Extract/Place the contents directly into the backend/ directory. This ensures the ML models and datasets are correctly located for the application to function.

Running the Application

Start the Backend Server

cd backend
python app.py

The application will start on http://127.0.0.1:5000.

Navigation

Route	Feature
/	Home / Dashboard
/ai-detection	AI Image Detection
/phishing	Phishing Detector
/url-safety	URL Safety Checker
/password	Password Toolkit
/awareness	Cyber Awareness Guide

Project Structure

NetShield/
├── backend/            # Flask application & ML models
│   ├── app.py          # Main entry point
│   ├── models/         # Trained models (pkl, pth)
│   ├── templates/ 
|   ├── requirements.txt    # Python dependencies
│   └── ...
├── frontend/           # Static assets & HTML pages
│   ├── assets/         # Images, icons, logos
│   ├── pages/          # Feature HTML pages
│   └── ...
└── README.md           # Documentation