skills/dp-archive/archive/spreadsheet/Gen Agent Trust Hub

spreadsheet

Pass

Audited by Gen Agent Trust Hub on Mar 6, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious intent, obfuscation, or unauthorized activities were detected. The skill follows established patterns for file processing and data analysis.
  • [COMMAND_EXECUTION]: The skill instructs the agent to use soffice (LibreOffice) and pdftoppm (Poppler) to render spreadsheets into visual formats for user verification. This is a primary feature of the skill and is restricted to its documented purpose.
  • [EXTERNAL_DOWNLOADS]: The documentation suggests installing standard, widely-used Python packages (openpyxl, pandas, matplotlib) and system utilities required for its rendering functionality.
  • [PROMPT_INJECTION]: The skill processes untrusted data from external spreadsheet files, which is a known surface for indirect prompt injection.
  • Ingestion points: openpyxl.load_workbook (seen in read_existing_spreadsheet.py) and various pandas loading functions referenced in SKILL.md.
  • Boundary markers: None; the skill does not explicitly implement delimiters to separate data from instructions within the spreadsheets.
  • Capability inventory: File system access (read/write) and system command execution for file rendering.
  • Sanitization: No sanitization or validation of spreadsheet content is performed before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 6, 2026, 01:57 PM