bioservices

Pass

Audited by Gen Agent Trust Hub on Apr 16, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill facilitates the installation of the 'bioservices' Python package from standard package registries.
  • [COMMAND_EXECUTION]: The provided scripts perform filesystem operations to manage biological data:
  • 'batch_id_converter.py', 'compound_cross_reference.py', and 'pathway_analysis.py' execute file writes to save analysis results in CSV and SIF formats.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes data retrieved from external bioinformatics web services and local user-provided files.
  • Ingestion points: External data enters the agent's context through API responses in 'protein_analysis_workflow.py' (UniProt, NCBI, KEGG) and 'compound_cross_reference.py' (KEGG, ChEMBL), as well as through local input files in 'batch_id_converter.py'.
  • Boundary markers: The skill does not implement delimiters or specific instructions to separate external data from agent commands.
  • Capability inventory: All Python scripts in the 'scripts/' directory perform network operations to scientific databases and/or conduct local file system writes.
  • Sanitization: There is no evidence of sanitization or filtering of the retrieved biological data to remove potential instructions before it is presented to the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 16, 2026, 10:49 PM