hunt-ntlm-info

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). The document is explicit offensive reconnaissance tooling: it provides ready-to-run probes and parsing code to extract NTLM Type‑2 AV_PAIRS (AD domain/tree names, hostnames, timestamps, UPN formats) and directly instructs chaining that intelligence into credential‑spray, hash‑cracking, and other attacks, which is high‑risk malicious guidance.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime workflow is to probe attacker-supplied internet endpoints (e.g., arbitrary https://target.example/... URLs) and then parse the server’s WWW-Authenticate: NTLM <base64> Type-2 challenge from the HTTP response headers, which is outsider-authored free text (remote server–generated content) injected into the agent’s LLM context via the captured response/decoded AV_PAIRS.