script-execute

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This skill intentionally accepts and compiles arbitrary C# source provided by callers (via the csharpCode input) and executes it with host privileges, enabling remote code execution that can be used for data exfiltration, credential theft, or to install backdoors if misused.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). The skill script-execute ingests outsider-authored free text via the csharpCode input at runtime (and compiles/executes it), making indirect prompt-injection-style payloads possible through the code string itself.