interpreting-routine-labs
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides structured clinical guidelines for laboratory interpretation and contains no executable code or malicious instructions.
- [NO_CODE]: Analysis of the skill file confirms that it contains no scripts, binaries, or external package dependencies.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external patient data (lab reports, medication lists), which represents a theoretical ingestion surface. However, because the skill has no configured tools for network access, file system modifications, or command execution, it lacks the necessary capabilities to be exploited by malicious data.
Audit Metadata