pharmaceutical-expert
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted external data related to clinical trials, FDA compliance, and pharmacovigilance, which can be used as a vector for indirect prompt injection attacks.
- Ingestion points: The skill is triggered by user-supplied information regarding clinical trials, FDA regulations, GxP standards, and drug development (identified in
SKILL.md). - Boundary markers: The instructions do not define clear delimiters or provided specific 'ignore internal instructions' directives when processing external pharmaceutical data or trial results.
- Capability inventory: The skill has access to powerful tools including
Read,Write,Bash, andWebSearchas defined in the YAML frontmatter ofSKILL.md. - Sanitization: There is no evidence of sanitization, validation, or schema-enforcement logic to filter malicious instructions embedded within clinical data or regulatory documents.
Audit Metadata