I need a set of modular Python scripts that bring machine-learning intelligence into my network-monitoring workflow. The primary goal is to spot malware traffic in real time, analyse it on the fly and trigger an automated response that contains or blocks the threat without manual intervention. Scope of work The solution should ingest live packet data (pcap, NetFlow or similar), run it through an AI/ML model trained to recognise malware signatures and behavioural anomalies, and then decide—within seconds—whether to raise an alert, quarantine the source or execute a predefined remediation playbook. Beyond simple flagging, I also want deeper malware analysis so I can review payload characteristics and produce short forensic reports. Required deliverables • Clean, well-documented Python scripts covering – malware analysis (static and/or dynamic) – real-time malware monitoring across the network – fully automated reaction once a threat is confirmed • A lightweight training or inference model (TensorFlow, PyTorch or scikit-learn are all fine) integrated into the scripts • Configuration file(s) so thresholds, network taps and response actions can be adjusted easily • A brief README with setup instructions and an example dataset for validation Acceptance criteria The system must detect and react to known and previously unseen malware samples in a controlled test network with fewer than 1 % false positives, generate a log entry with timestamp, source, destination and action taken, and run on a standard Linux server with no proprietary dependencies. If you are comfortable combining Python, packet capture tools like Wireshark/tshark and modern ML libraries to build a dependable, self-healing security layer, let’s get started.