Selected technical engineering projects covering Model Context Protocol (MCP) servers for autonomous AI security researchers, adversarial machine learning games, physical IoT honeypot sensor networks, and red-team penetration testing tooling.
Custom Model Context Protocol (MCP) server implementation built in Python. Empowers AI coding assistants and autonomous agents to interface directly with daily security tools (Nmap port scanning, CVE search, Shodan queries, Hash identifiers) via structured JSON-RPC APIs.
An interactive educational cybersecurity game built in TypeScript. Simplifies complex neural network internals, gradient updates, prompt injection vulnerabilities, and adversarial model evasion into a gamified challenge format.
Dual-channel hardware intrusion decoy operating on internal subnets. Listens for unauthorized ARP broadcasts and network sweeps, immediately dispatching tamper notifications via Telegram Bot API and TLS-encrypted Gmail alerts.
Modular penetration testing toolkit for enterprise Active Directory environments. Automates Kerberoasting ticket requests, AS-REP roasting triage, BloodHound graph path mapping, and credential extraction verification.
Public archive of practical penetration testing walkthroughs on vulnerable HackMyVM challenge machines. Deep-dives into privilege escalation, custom SUID binary reverse engineering, and web shell payloads.
Automated network forensic pipeline ingesting raw PCAPs, matching malicious traffic against Suricata alert feeds, fingerprinting JA3/JA3S client hashes, and extracting malware indicators of compromise (IOCs).
Try searching for keywords like "MCP", "Neuroquest", "HoneyBot", "Active Directory", or select "All Projects".