best-of-n
Pass
Audited by Gen Agent Trust Hub on Nov 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill establishes an orchestration pattern where untrusted candidate outputs from workers are processed by a checker agent. This represents an attack surface for indirect prompt injection.
- Ingestion points: Candidate outputs are stored in a shared-VM directory (SKILL.md).
- Boundary markers: No delimiters or protective instructions are defined to isolate untrusted content.
- Capability inventory: The skill manages agent communication via the group-member capability but does not define shell commands, file system writes, or network operations.
- Sanitization: No sanitization or validation of candidate output is specified.
Audit Metadata