Technical Specifications.
Verified parameters and operational benchmarks for this release.
Tech Stack.
Features Checklist.
Core capabilities, verified deliverables, and technical assets included with this release.
- Cyber-physical system simulation
- Synthetic multi-agent CPS scenario generation
- Malware/anomaly detection
- Graph-temporal anomaly scoring
- PyTorch graph-temporal neural model
- LSTM-based temporal modeling
- Synthetic labeled training dataset generation
- Model training pipeline
- Model evaluation and metrics
- Training artifact persistence
- Real-time alert generation
- Threat severity tracking
- Trust-score reduction
- Agent isolation
- Safe-mode mitigation
- Automated mitigation policy engine
- SQLite alert persistence
- Simulation run persistence
- REST API
- Training API
- Minimal security monitoring dashboard
- Configurable CPS simulation
System Requirements.
- OS
- Linux, Windows, or macOS
- CPU
- 4+ cores recommended
- GPU
- NVIDIA GPU with 8GB+ VRAM recommended for model training
- RAM
- 8GB minimum, 16GB recommended
- Python
- 3.10+
- Browser
- Modern Chrome, Firefox, Edge, or Chromium-based browser
- Network
- Localhost or LAN deployment
- Storage
- 5GB+ free space
About this Project.
AI Malware Detection & Mitigation Platform for Cyber-Physical Systems
An end-to-end cybersecurity platform for simulated cyber-physical systems that combines intelligent-agent simulation, graph-temporal anomaly detection, machine learning, automated mitigation, and security monitoring.
The platform generates synthetic multi-agent CPS scenarios and analyzes system behavior to identify potentially malicious or anomalous activity. Detected threats can trigger mitigation actions including trust reduction, agent isolation, and safe-mode transitions.
Core Components
CPS Simulation
Uses SimPy to generate synthetic cyber-physical system scenarios involving multiple interacting agents.
AI Detection
Includes a baseline graph-temporal anomaly detector and a PyTorch training pipeline for developing graph + temporal malware detection models.
Mitigation Engine
Provides policy-driven responses to detected threats, including:
Trust reduction Agent isolation Safe-mode activation Alert generation
Machine Learning
Includes synthetic dataset generation, model training, evaluation, and artifact persistence.
Security Dashboard
A lightweight dashboard provides visibility into simulation runs, alerts, detection results, and system state.
REST API
The FastAPI backend exposes APIs for simulation, detection, training, metrics, and system monitoring.
Training
The platform can generate labeled synthetic data and train the included baseline graph-temporal model through the training API.
Included Complete source code FastAPI backend SQLite persistence CPS simulation engine Malware/anomaly detection service Graph-temporal PyTorch model Training pipeline Synthetic dataset generator Automated mitigation engine Security dashboard API endpoints Setup documentation Intended Use
Suitable for:
Cybersecurity research AI/ML security projects Cyber-physical system research Industrial cybersecurity prototypes Malware detection research Intelligent-agent security experiments Academic demonstrations Proof-of-concept development Important
This implementation is designed as a research/prototyping platform and uses synthetic CPS scenarios. It should not be treated as a production industrial-control security system without additional validation, authentication, hardening, real-world datasets, and deployment testing.