salah9003/Automated-Vulnerability-Scanning-with-Agentic-AI

Multi-agent architecture with specialized roles (strategy generation, command execution, error handling, reporting) coordinated via GPT-4 Turbo to autonomously scan targets on Kali Linux using standard penetration testing tools. The system iteratively refines scanning strategies based on command output feedback until completion, with all interactions and findings logged to markdown reports. Designed as a proof-of-concept demonstrating LLM-based agent collaboration for cybersecurity workflows, though interactive tools like msfconsole have execution limitations.

No commits in the last 6 months.