libsbml-network-modeling
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches biological model files from the BioModels Database, which is an established and well-known scientific service provided by the European Bioinformatics Institute (EBI). This is documented neutrally as it targets a well-known research domain.
- [INDIRECT_PROMPT_INJECTION]: The skill includes functionality to ingest SBML XML files from external sources and local storage, which creates a surface where data could theoretically contain instructions intended to influence the agent.
- Ingestion points: External data enters the context through
urllib.request.urlretrieveand is processed usinglibsbml.SBMLReader().readSBMLFromFile(). - Boundary markers: There are no explicit delimiters or specific warnings to ignore instructions embedded within the SBML XML fields.
- Capability inventory: The skill possesses capabilities for file system write operations (
SBMLWriter().writeSBMLToFile()) and network retrieval (urllib.request.urlretrieve). - Sanitization: The skill performs schema validation using the
libsbmlAPI to ensure SBML specification compliance, though it does not implement specific sanitization of textual fields for prompt injection patterns.
Audit Metadata