skills/jimliuxinghai/skills/xlsx/Gen Agent Trust Hub

xlsx

Pass

Audited by Gen Agent Trust Hub on Mar 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The recalc.py script utilizes subprocess.run to invoke the soffice (LibreOffice) binary for formula evaluation, which is a necessary step for dynamic spreadsheet generation.
  • [COMMAND_EXECUTION]: The recalc.py script modifies the user's environment by writing a StarBasic macro file (Module1.xba) to the LibreOffice configuration directory to enable automated calculation and saving.
  • [COMMAND_EXECUTION]: The script checks for and utilizes system timeout or gtimeout utilities to ensure the headless LibreOffice process does not exceed defined execution limits.
  • [PROMPT_INJECTION]: The skill processes untrusted spreadsheet data, presenting a surface for indirect prompt injection. Ingestion points: Data is loaded via pd.read_excel and load_workbook as described in SKILL.md. Boundary markers: The agent is instructed to follow existing template conventions and use provided validation scripts. Capability inventory: recalc.py performs subprocess execution and file system writes; pandas and openpyxl perform file I/O. Sanitization: The recalc.py script includes a post-calculation scanning phase that identifies common Excel error codes (e.g., #REF!, #VALUE!) to ensure model integrity.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 13, 2026, 01:49 AM