Active Directory Attacks

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt repeatedly embeds plaintext credentials, hashes, and secret placeholders directly into command examples (e.g., -u 'user' -p 'password', domain/user:password, admin:password, KRBTGT_HASH, NTHASH), which requires the LLM to handle or output secret values verbatim and therefore poses high exfiltration risk.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This document is an explicit offensive playbook containing step‑by‑step instructions and tooling to steal credentials (Kerberoast, AS‑REP, DCSync, Mimikatz), forge/plant Kerberos tickets (Golden/Silver Tickets), perform NTLM relays and certificate/ADCS abuses, and deploy persistent backdoors (via SCCM/WSUS, GPO abuse, RBCD), which together constitute deliberate malicious activity and supply‑chain/persistence capabilities.

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 system-changing commands (e.g., "sudo date -s" to change system time) and contains multiple offensive actions that modify remote/host state (credential extraction, DCSync, Golden Ticket creation, restoring DC passwords) which push the agent to perform privileged or state-changing operations.