document-xlsx
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external, untrusted spreadsheet files and structured data (CSV, JSON, Parquet), which inherently exposes an attack surface for indirect prompt injection or malicious Excel formulas.
- Ingestion points:
scripts/xlsx_audit.py,scripts/xlsx_export_report.py, andscripts/xlsx_sanitize.pyserve as the primary entry points for external data processing. - Boundary markers: The skill explicitly instructs the agent to treat external workbooks as hostile and provides a checklist for security reviews.
- Capability inventory: The skill uses file system access and shell command execution to run Python-based auditing and export tools.
- Sanitization: The
scripts/xlsx_sanitize.pytool provides proactive protection by identifying and prepending a single quote to dangerous formula-triggering characters (=,+,-,@) in text strings, ensuring they are treated as literal text rather than executable formulas.
Audit Metadata