ln-72-product-outcome-evaluator
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external product data and customer behavior evidence which are potentially untrusted sources.
- Ingestion points: Untrusted data enters the agent context via 'authorized analytics' and 'experiment results' as defined in the Tool Routing section.
- Boundary markers: The instructions lack explicit boundary markers or delimiters to separate external data from system instructions.
- Capability inventory: The skill is strictly read-only and lacks capabilities for file modification, network communication, or system command execution.
- Sanitization: There are no requirements for escaping or sanitizing ingested evidence before processing.
Audit Metadata