ai-security

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill includes instructions that explicitly request verbatim reproduction of system prompts and shows an inline API-key-like string and an exfiltration URL pattern, which would require the LLM to handle or output secret values directly.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). The document contains explicit, actionable examples of prompt injection, secret extraction, data exfiltration, and remote code execution (e.g., pickle execution and malicious model code), which are high-risk if reused maliciously.