skills/ynulihao/agentskillos/xlsx/Gen Agent Trust Hub

xlsx

Pass

Audited by Gen Agent Trust Hub on Sep 19, 2026

Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONPERSISTENCEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The recalc.py script executes local system commands using subprocess.run to trigger formula recalculation in Excel files via LibreOffice.\n- [DYNAMIC_EXECUTION]: The skill dynamically generates a LibreOffice StarBasic macro at runtime from a hardcoded template and saves it to the filesystem to enable automated recalculation features.\n- [PERSISTENCE]: The recalc.py script installs a persistent macro file in the user's LibreOffice configuration directory (~/.config/libreoffice/ or ~/Library/Application Support/LibreOffice/), which is required for the tool to function across sessions.\n- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest data from external Excel files, creating an inherent attack surface for malicious instructions embedded in spreadsheet cells.\n
  • Ingestion points: Use of pandas.read_excel and openpyxl.load_workbook examples in SKILL.md and recalc.py.\n
  • Boundary markers: No explicit delimiters or instructions to ignore embedded content were identified.\n
  • Capability inventory: The skill has capabilities to execute shell commands via subprocess.run and perform file system read/write operations.\n
  • Sanitization: No sanitization of spreadsheet cell content is performed before processing or display.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 19, 2026, 12:32 PM