🛡 Keylogger Detection System

Multi-layer detection of malicious keyloggers combining:

• 🖥 Real-time process & file monitoring
• 🐍 Simulated keylogger for adversarial testing
• 🤖 Machine Learning classification (LogReg, Random Forest, Gradient Boosting)
• 🌐 Web-scraped process dataset for training

---

🚀 How It Works

1. Heuristic Monitor — watches running processes, suspicious file creation, and clipboard anomalies (keylogger/py1).
2. Simulated Keylogger — generates malicious behavior for testing (keylogger/py2).
3. Machine Learning — trains models on labeled process data (keylogger_ml/ml_integration.py).
4. Data Collection — captures process snapshots & scrapes processlibrary.com for benign samples.

---

📂 Project Structure

keylogger/         # Heuristic monitor & simulated keylogger
keylogger_ml/      # ML training & evaluation pipeline
task_manager.py    # Process snapshot tool
web_scraping.py    # Process name scraper
process_data.csv   # Labeled dataset (malicious / benign)

---

🛠 Tech Stack

Python, psutil, pynput, cryptography, scikit-learn

Data Handling: Pandas, NumPy

Visualization: Matplotlib, Seaborn

Scraping: BeautifulSoup4, lxml

Windows APIs: win32clipboard, ImageGrab

🎯 Future Goals

Integrate trained model into live monitor for real-time detection

Expand features beyond process names (e.g., behavioral metrics)

Add model persistence & dashboard alerts

---

Disclaimer: This project includes a simulated keylogger component for educational and research purposes only. Do not deploy on systems without explicit permission.