ai-llm-app-attack

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This document explicitly describes and instructs on multiple exploitation techniques (prompt injection, SSRF, RCE, webshell writing, SQL exfiltration, model-pickle RCE), indicating clear malicious intent and actionable abuse patterns.

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

• Attempt to modify system services in skill instructions detected (high risk: 1.00). Flagged because the prompt explicitly describes and encourages actions that read sensitive files (e.g., /etc/passwd), write webshells, perform SSRF to cloud metadata, and other tool-driven RCE/side-effect attacks that would modify or exfiltrate the host state.