pyopenms

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill consists of legitimate Python scripts and reference documentation for scientific research using the pyOpenMS library. All scripts are parameterized CLI tools that perform intended scientific workflows.
  • [EXTERNAL_DOWNLOADS]: The documentation refers to downloading an optional chemical structure mapping file (HMDB2StructMapping.tsv) from the official OpenMS GitHub repository. This is a well-known and trusted source for mass spectrometry software and data.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted mass spectrometry data files, which presents a theoretical injection surface common to data processing tools.
  • Ingestion points: Input arguments in scripts such as inspect_ms_data.py and detect_features_metabo.py accept mzML, mzXML, and featureXML files.
  • Boundary markers: None present, as these data formats are structured scientific schemas rather than natural language instructions.
  • Capability inventory: Includes file system write access for exporting results (CSV, XML, PNG) and computational mass spectrometry algorithms.
  • Sanitization: Relies on the internal parsing and validation logic of the pyOpenMS library to handle specific data schemas.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 01:47 AM
Security Audit — agent-trust-hub — pyopenms