cmdi-command-injection

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.95). The list contains multiple high-risk indicators — attacker-controlled domains and Burp Collaborator placeholders for OOB exfiltration, a direct PowerShell script URL (shell.ps1), and an exploit GitHub repository — which are classic vectors for distributing/mounting malware (while a few entries are benign placeholders like example.com or 127.0.0.1, the overall set is suspicious).

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This document is an explicit offensive command‑injection playbook containing step‑by‑step payloads and techniques (OOB DNS/HTTP exfiltration, reverse shells, remote code execution, persistence and privilege escalation, and obfuscation/WAF‑bypass methods) intended to enable unauthorized data theft and system compromise — clearly malicious.

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

• Attempt to modify system services in skill instructions detected (high risk: 1.00). The prompt is an explicit command-injection attack playbook containing payloads and instructions (reverse shells, reading/writing /etc files, writing CGI/.htaccess, LD_PRELOAD, OOB exfiltration and privilege-escalation techniques) that direct actions which modify system state and bypass security controls.