xlsx-custom

Pass

Audited by Gen Agent Trust Hub on Jul 9, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The script scripts/recalc.py invokes the soffice or libreoffice binary using subprocess.run() to recalculate spreadsheet formulas in headless mode.
  • [COMMAND_EXECUTION]: The scripts/recalc.py script uses importlib to dynamically load and execute a Python module from a path computed at runtime (office-custom/scripts/soffice.py), which involves dynamic execution of code from computed filesystem paths.
  • [PROMPT_INJECTION]: The skill processes untrusted spreadsheet data and possesses capabilities to write files and execute system commands, presenting a surface for indirect prompt injection.
  • Ingestion points: Files loaded via pandas.read_excel() and openpyxl.load_workbook() as described in SKILL.md and scripts/recalc.py.
  • Boundary markers: None identified in the skill instructions or scripts.
  • Capability inventory: File writing (wb.save, df.to_excel), complex data manipulation (pandas), and system command execution (soffice via recalc.py).
  • Sanitization: No validation or sanitization of data extracted from spreadsheet cells is implemented before further processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 9, 2026, 03:00 AM
Security Audit — agent-trust-hub — xlsx-custom