tooluniverse-pharmacogenomics
Warn
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill explicitly directs the agent to use Bash to run Python scripts for data processing and statistical computation rather than just describing results.
- [DYNAMIC_EXECUTION]: The instructions mandate the runtime generation and execution of Python code (utilizing libraries such as pandas and scipy) to perform data analysis and enrichment based on task requirements.
- [INDIRECT_PROMPT_INJECTION]: The skill presents an attack surface for indirect prompt injection as it processes data retrieved from numerous external pharmacogenomic sources (e.g., PharmGKB, CPIC, FDA, DGIdb) without defining boundary markers or implementing sanitization logic to prevent embedded instructions in the fetched data from influencing the agent's behavior or generated code.
Audit Metadata