Pentest Commands

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This skill content contains explicit, actionable offensive commands (Metasploit exploits and msfvenom reverse shells, sqlmap --os-shell, brute-force tools like Hydra/John, Wi‑Fi deauth/aircrack, network exploitation and enumeration) that enable backdoors/remote code execution, credential theft, and network/system compromise if used without authorization.

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 supplies explicit, actionable commands for launching exploits, brute-forcing, capturing traffic and running tools that require root (e.g., airmon-ng, tshark, msfvenom/msfconsole, sqlmap --os-shell), which would cause the agent's host to change network/interface state, open listeners and write/execute payloads—so it pushes potentially state-changing, privileged operations even if it doesn't directly instruct creating users or editing system files.