sn-da-non-spreadsheet-analysis

Pass

Audited by Gen Agent Trust Hub on Jul 3, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses subprocess.run to call libreoffice for converting legacy document formats (.doc, .ppt) to modern ones (.docx, .pptx). These calls use argument lists rather than shell strings, which prevents shell-based command injection.
  • [COMMAND_EXECUTION]: It invokes a local utility script caption.py from a related skill to perform OCR and image analysis on document pages lacking extractable text. The execution is structured to prevent argument-based injection.
  • [SAFE]: File access is restricted to user-provided data directories and temporary storage (/tmp/), which is standard for file processing and data analysis workloads.
  • [SAFE]: The skill correctly implements directory scanning and multi-file processing, ensuring that user requests for batch analysis are handled completely without stopping at the first file.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 3, 2026, 12:24 PM
Security Audit — agent-trust-hub — sn-da-non-spreadsheet-analysis