abusing-shadow-credentials-for-privesc

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill prompt explicitly shows and requires embedding secret values verbatim in commands and tool invocations (e.g., -p "Passw0rd!", <PFX_PASSWORD>, , ), which forces the LLM to handle and output sensitive credentials directly, creating an exfiltration risk.

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.90). These links include GitHub repositories (Whisker, pyWhisker, Certipy, PKINITtools) and how‑to documentation for the Shadow Credentials AD takeover technique—legitimate security‑research resources hosted on reputable sites but they distribute offensive tooling that can perform stealthy credential theft/privilege escalation and are high risk if downloaded or run outside an authorized engagement.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). The content is an explicit, dual‑use offensive playbook and tooling orchestration that deliberately modifies Active Directory msDS-KeyCredentialLink to inject attacker-controlled credentials, obtain PKINIT TGTs, extract NT hashes, and enable account takeover and lateral movement (credential theft and backdoor-style access).

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 modifying Active Directory objects (msDS-KeyCredentialLink), adding attacker-controlled credentials, and performing account takeover (NT hash extraction/pass-the-hash), which are direct, state-changing, privileged operations that compromise target machines and domain state.