do
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns, obfuscation, or unauthorized access attempts were detected. The skill's primary function is to provide a clean execution context for subtasks.
- [PROMPT_INJECTION]: The skill uses the
$ARGUMENTSplaceholder to include user input in its instructions. While it lacks explicit boundary markers to isolate this input, the skill's configuration significantly restricts tool access (defined as an empty string in the frontmatter), which effectively mitigates the risk of an indirect prompt injection attack attempting to exploit the execution environment.
Audit Metadata