cobrapy
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill facilitates the retrieval of metabolic model files from external scientific repositories, specifically the BiGG Models database (hosted by the University of California, San Diego) and the BioModels repository. These operations use Python's
urllib.requestwith security-conscious implementations such as timeouts and file size limits.\n- [INDIRECT_PROMPT_INJECTION]: The processing of external model definitions in SBML, JSON, and YAML formats represents an ingestion surface for indirect prompt injection attacks.\n - Ingestion points: Data enters the agent context via
read_sbml_model,load_json_model, andload_modelfunctions as documented inreferences/api_quick_reference.md.\n - Boundary markers: The skill instructions do not provide explicit delimiters or warnings to ignore instructions embedded within the metadata or annotations of these ingested files.\n
- Capability inventory: The agent has access to
Bash,Write, andEdittools, and executes Python code for modeling, which increases the potential impact of a successful injection.\n - Sanitization: The skill lacks mechanisms to sanitize or validate non-structural fields (such as reaction names or metabolite descriptions) before they are presented to the agent or user.
Audit Metadata