ssti-twig

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt explicitly directs the agent to discover, record, and report "new credentials or tokens found" and to save evidence (e.g., wp-config.php, ssrf-aws-creds.json), which requires outputting secret values verbatim and thus creates a 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 guide for exploiting server‑side template injection (Twig/Smarty/Blade/Latte) to achieve remote code execution, read sensitive files and credentials, exfiltrate data, write webshells, and bypass protections using obfuscation — it contains multiple high‑risk, clearly malicious techniques.

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

• Attempt to modify system services in skill instructions detected (high risk: 0.80). The skill explicitly instructs exploiting SSTI to achieve server-side remote code execution and file reads/writes (e.g., writing webshells, reading/writing /etc/passwd, using system/exec), which encourages modifying the target host's filesystem and running arbitrary commands and thus can compromise the machine state (and potentially the agent host if targeted locally).