performing-directory-traversal-testing

SUSPICIOUS: the skill is internally coherent as a penetration-testing playbook, but its actual footprint is an offensive exploit workflow that teaches an AI agent to read credential files and pursue LFI-to-RCE on remote targets. Install trust is mostly acceptable, yet the exploitation and sensitive-file retrieval capabilities make the overall skill high risk for agent deployment.