interpreting-tumor-markers
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill consists entirely of markdown-based instructional content and diagnostic frameworks. It contains no scripts, commands, or binary files that could be executed by the agent environment.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and interpret external medical data. 1. Ingestion points: Lab results, pathology reports, and treatment timelines (Checkpoint A). 2. Boundary markers: The process is gated by Checkpoints A and B. 3. Capability inventory: The skill specifies no active tools, file-writing, or network capabilities. 4. Sanitization: No explicit data sanitization is performed. Because the skill lacks functional capabilities to abuse, the data ingestion remains safe.
Audit Metadata