password-spraying

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs generating and executing scripts that include and echo plaintext passwords (e.g., re-test commands with 'HITPASS', reporting "New credentials" and saving valid credentials to files), so the LLM must handle and output secret credential values verbatim.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This document is an explicit how-to for password spraying, credential harvesting, lateral movement, and remote authentication abuse (including remote command execution and account takeover), and includes instructions to collect and report harvested credentials — behaviors that directly enable unauthorized access and data exfiltration if misused.

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 the agent to disable the execution sandbox (dangerouslyDisableSandbox=true) and to run privileged commands (e.g., "sudo ntpdate DC_IP"/"sudo rdate"), which asks the agent to bypass security mechanisms and obtain root to modify the host state.