diagram
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data via the
$ARGUMENTSvariable and includes a directive for the agent to save the resulting diagrams to files in the project. This combination lacks boundary markers or sanitization, creating a surface where malicious instructions within the arguments could lead the agent to write unintended or malicious data to the project's filesystem. - Ingestion points:
$ARGUMENTSplaceholder inSKILL.md. - Boundary markers: None present to delimit instructions from user-provided data.
- Capability inventory: Instruction to write to files within the project codebase.
- Sanitization: No validation or escaping of the generated diagram content before it is saved to a file.
Audit Metadata