expert-naval
Pass
Audited by Gen Agent Trust Hub on Nov 2, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill suggests loading itself and its dependencies via the
npx openskillscommand. This is a standard platform mechanism for managing expert modules and does not constitute a malicious operation. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to process user-provided career and life scenarios, which is a routine surface for indirect prompt injection where an attacker might try to override the agent's persona.
- Ingestion points: Untrusted data enters the agent context through user questions about life and career (e.g., 'How do I make more money?').
- Boundary markers: The skill does not implement specific delimiters or instructions to ignore instructions embedded in user input.
- Capability inventory: The skill provides purely informational frameworks and mental models; it does not define or request access to high-risk tools or system capabilities.
- Sanitization: There is no evidence of input validation or sanitization logic in the provided files.
Audit Metadata