path-traversal-lfi

This skill is internally consistent as an exploit playbook, but its footprint is a high-risk offensive-security capability set for an AI agent. It does not show hidden exfiltration or deceptive install behavior, yet it meaningfully enables file theft, credential discovery, remote payload delivery, and command execution against targets; classify as suspicious/high-risk vulnerable, not confirmed malware.