agency-penetration-tester

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This document contains explicit, actionable offensive instructions (reverse shells, C2/tunneling, credential theft, DCSync/Golden Ticket, persistence/backdoors, and data exfiltration techniques) that can be directly abused for unauthorized compromise and data theft.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 0.90). The skill explicitly instructs running privileged host-altering commands (e.g., "sudo ip route add ...") and tooling (meterpreter autoroute, proxy setup) that modify network routing/state on the operator's machine, which requires elevated privileges and can compromise the agent host.