drug-safety-specialist
Pass
Audited by Gen Agent Trust Hub on Apr 21, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: No malicious content, obfuscation, or unauthorized commands were detected. The skill contains documentation and persona instructions without any executable scripts or tool-access configurations. All external references target reputable regulatory bodies such as the FDA, EMA, and ICH.
- [PROMPT_INJECTION]: The skill processes untrusted external data (adverse event reports, medical literature), which is a surface for indirect prompt injection. However, since the skill is a persona with no tool-use capabilities (no file system or network tools), the risk is negligible and categorized as safe. Ingestion points include adverse event reports, literature monitoring, and clinical trial data. Boundary markers are not provided in the prompt templates. Capability inventory shows no tools or subprocess execution capability. Sanitization is not mentioned.
- [NO_CODE]: The skill does not include any executable scripts, binaries, or configuration files that run code, significantly reducing the potential attack surface.
Audit Metadata