bioservices
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests scientific identifiers from user input and local files to query 40+ bioinformatics services, creating a surface for indirect prompt injection.
- Ingestion points: Multiple scripts including
batch_id_converter.py(file input) andprotein_analysis_workflow.py(CLI arguments) accept external identifiers. - Boundary markers: The instructions and scripts do not utilize specific delimiters to isolate user-provided data from agent instructions.
- Capability inventory: The skill has the ability to perform network queries via the bioservices Python library and write results to local storage.
- Sanitization: Identifiers are passed to API endpoints without comprehensive sanitization or schema validation.
Audit Metadata