pyopenms

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • External Downloads (SAFE): The skill installs the pyopenms package via uv. While the authoring organization 'OpenMS' is not in the predefined trusted list, it is a well-recognized scientific project. Dependencies include standard data science packages like pandas, numpy, and scikit-learn.\n- Indirect Prompt Injection (LOW): The skill ingests untrusted data from various external mass spectrometry file formats.\n
  • Ingestion points: Multiple files including SKILL.md and references/file_io.md demonstrate loading data via ms.MzMLFile().load(), ms.FASTAFile().load(), and ms.IdXMLFile().load().\n
  • Boundary markers: Absent. There are no instructions to the agent to ignore potentially malicious instructions embedded in data files.\n
  • Capability inventory: The skill includes complex data processing but lacks high-risk capabilities like arbitrary command execution or network exfiltration.\n
  • Sanitization: No sanitization of the input file content is performed before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 05:01 PM