ntlm-relay-coercion

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The skill includes literal password strings and examples of passing passwords on the command line (e.g., "P@ss123" and "-p pass"), which requires the LLM to emit credential values verbatim and creates an exfiltration risk.

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.70). The GitHub URL points to an offensive tooling repository (Coercer) that can be used to perform authentication coercion and attacks and is therefore potentially risky to download/run from untrusted sources; the CA_HOST entry is just an internal AD CS endpoint placeholder and not a public download link.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This content is a deliberate offensive playbook providing step-by-step instructions and tooling to capture credentials, perform NTLM coercion/relay, obtain unauthorized access (RBCD/shadow creds/ADCS certificates), and achieve remote code execution and domain compromise.

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 editing a system config file (/etc/responder/Responder.conf) and running privileged network/native tools (Responder, mitm6, ntlmrelayx) and even adding computer accounts on a domain (ntlmrelayx --add-computer), which require elevated privileges and modify system or remote state.