pymatgen
Pass
Audited by Gen Agent Trust Hub on Nov 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from various materials science file formats (e.g., CIF, POSCAR, XYZ) which could theoretically contain malicious instructions in metadata or comments.
- Ingestion points: The skill uses
Structure.from_file()inscripts/structure_analyzer.pyandscripts/structure_converter.pyto read external data files. - Boundary markers: There are no explicit delimiters or system instructions to ignore potential natural language commands embedded within the processed data files.
- Capability inventory: The skill has the capability to perform network requests (to the Materials Project API) and write files to the local file system (during format conversion).
- Sanitization: Data is parsed using the standard pymatgen library, but no additional sanitization is applied to text-based fields before they are presented to the agent's context.
- [EXTERNAL_DOWNLOADS]: The skill documentation recommends installing standard scientific packages (
pymatgen,mp-api) and utilizes the Materials Project API. These are well-known and trusted resources within the computational materials science community.
Audit Metadata