document-xlsx

Pass

Audited by Gen Agent Trust Hub on Oct 5, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses a data ingestion surface as it is designed to process external spreadsheet data (XLSX, CSV, JSON, Parquet) via scripts such as scripts/xlsx_audit.py and scripts/xlsx_export_report.py.
  • Ingestion points: Reading of local data files for conversion or audit.
  • Boundary markers: Includes assets/spreadsheet-model-review-checklist.md to guide manual review of workbook logic.
  • Capability inventory: Capability to read and write files and execute Bash commands for automation.
  • Sanitization: The skill contains proactive security measures, specifically scripts/xlsx_sanitize.py, which is designed to quote dangerous formula prefixes (e.g., =, +, -, @) to mitigate CSV and formula injection attacks.
  • [EXTERNAL_DOWNLOADS]: The skill references documentation and official GitHub repositories for standard technology stacks, including openpyxl, pandas, calamine, and IBM Research's Docling. These references target well-known organizations and official package registries and are documented neutrally as technical resources.
  • [COMMAND_EXECUTION]: The skill uses standard Python scripts for file manipulation. The scripts utilize benign built-in modules such as argparse, pathlib, and zipfile to handle OOXML (Excel) archives without employing dangerous dynamic execution patterns.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 5, 2026, 01:59 PM