deeper
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied tasks via the
$ARGUMENTSplaceholder. This introduces a surface for indirect prompt injection where a task could contain instructions to subvert the agent. However, the skill's primary function is to generate a rewritten prompt for user review, and it lacks dangerous capabilities like network access or file system modification. - Ingestion points:
$ARGUMENTSinSKILL.md. - Boundary markers: None present.
- Capability inventory: None (no subprocess calls, file writes, or network operations).
- Sanitization: No explicit sanitization or escaping of the input is performed.
- [NO_CODE]: The skill is composed entirely of markdown instructions and YAML metadata. It does not include any scripts (Python, JavaScript, shell) or external dependencies, significantly reducing the potential for malicious code execution.
Audit Metadata