arjun

The skill is internally consistent with its stated purpose, but that purpose is to equip an AI agent with offensive web-security testing workflows. Installation trust is relatively normal (PyPI, no curl|bash), yet the capability scope is high-risk because it enables authenticated scanning, passive recon, and direct handoff into exploit-oriented tools. Suspicious overall due to offensive-agent use, not because of hidden exfiltration or deceptive install behavior.