conducting-internal-network-penetration-test

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt includes numerous hardcoded plaintext credentials and directs using them verbatim in commands (e.g., -u/-p, connection strings, and report examples), requiring the LLM to handle and emit secret values directly.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This content provides explicit, actionable offensive capabilities (LLMNR/NTLM poisoning, NTLM relays, Mimikatz, DCSync/GoldenTicket, lateral movement with PsExec/WMI/WinRM, automated AD enumeration and credentials harvesting) that are clearly deliberate attack patterns and readily abuseable as backdoor/credential-theft and domain-compromise tooling.

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

• Attempt to modify system services in skill instructions detected (high risk: 1.00). The skill explicitly instructs running privileged and state-changing actions on the host (e.g., "sudo responder", "sudo mitm6", meterpreter getsystem, privilege-escalation tools, writing proof files and network-poisoning/relay services) which require/encourage sudo and modify network and system state, so it pushes the agent to compromise the machine.