AI Sentinel is a real-time, machine-learning-powered firewall designed for PCs. It uses an Isolation Forest model to monitor network traffic and detect anomalies, providing an interactive web dashboard for visualization and control.

• ML Detection Engine: High-performance classification using the trained firewall_anomaly_detection_model.pkl.
• Premium Dashboard: A sleek, dark-themed UI built with FastAPI, Tailwind CSS, and Chart.js.
• Real-Time Analytics:
  • Anomaly Risk Charts: Dynamic line charts plotting risk scores in real-time.
  • Traffic Stats: Detailed counters for total flows and detected anomalies.
  • Activity Logs: An auto-updating table of detailed network flow events.
• Interactive Controls:
  • Sensitivity Threshold: Adjust the model's sensitivity on-the-fly.
  • Simulation Mode: Safely test the firewall in a sandbox environment using the provided dataset.

• firewall_engine.py: The core classification and packet processing engine.
• dashboard_app.py: FastAPI server that serves the dashboard and manages the core engine.
• static/index.html: The premium dashboard frontend.
• firewall_anomaly_detection_model.pkl: The trained Isolation Forest model and scaler.
• optional/: Contains non-essential files, including log2.csv, ESP32 deployment code, and research notebooks.

Ensure you have Python 3.8+ installed. It is recommended to use the existing virtual environment.

Run the following in your terminal:

.\venv\Scripts\python.exe -m pip install fastapi uvicorn pandas scikit-learn numpy plyer pydivert

Option A: Manual Start (Visible)

python dashboard_app.py

Option B: Background Start (Hidden) Double-click start_firewall.vbs to launch the firewall invisibly.

To monitor real-time network traffic instead of simulation:

1. Open the dashboard at http://localhost:8000.
2. Toggle "Simulation Mode" to OFF.
3. Click "Start Engine".
4. Note: Real-time traffic monitoring requires running the dashboard/terminal as Administrator because it uses WinDivert to intercept packets.

To have the firewall turn on automatically with your PC:

1. Press Win + R, type shell:startup, and press Enter.
2. Create a Shortcut to start_firewall.vbs in that folder.
3. The AI Sentinel will now launch in the background every time you log in!

The firewall will now send Windows Toast notifications whenever a threat is detected by the ML model.

Built for the PC Firewall AI project.