establishing-persistence

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt includes plaintext credentials and keys embedded directly in commands and code examples (e.g., net user backdoor P@ssw0rd, echo "ssh-rsa AAAA...attacker_key", and hardcoded webshell secrets), which requires the LLM to output secret values verbatim.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This content is explicitly a how-to for creating persistent backdoors and long-term unauthorized access across Windows, Linux, containers, and cloud environments (registry Run keys, scheduled tasks, services, WMI subscriptions, cron/systemd/init scripts, SSH authorized_keys, AppInit_DLLs/LD_PRELOAD, kernel modules, web shells, IAM/service principals), and therefore clearly facilitates malicious post‑exploitation and system compromise.

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

• Attempt to modify system services in skill instructions detected (high risk: 1.00). This skill explicitly instructs creating backdoor accounts, modifying system and service files, adding registry keys, cron/systemd jobs, SSH keys, and rootkits—actions that change system state and require elevated privileges to establish unauthorized persistence.