proxychains-tunnel

SUSPICIOUS. The skill is internally consistent with its stated purpose, but that purpose is to equip an AI agent with tunneled offensive-security command execution. No malicious installer or credential-harvesting flow is shown, yet the skill enables high-impact recon against remote hosts and sends egress-check data to a third-party IP service. Risk is elevated mainly from agentized security-tool use, not from hidden malware behavior.