rice
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user data to generate RICE scores, which could theoretically contain instructions to influence agent output.
- Ingestion points: User-provided initiatives are ingested via the
$ARGUMENTSvariable at the end ofSKILL.md. - Boundary markers: There are no explicit delimiters or specific instructions provided to the agent to treat the
$ARGUMENTScontent as data only and to ignore any embedded instructions. - Capability inventory: No dangerous capabilities such as network operations, file system writes, or subprocess executions were identified in the skill.
- Sanitization: The skill does not perform any sanitization, validation, or escaping of the user-provided content before processing.
Audit Metadata