skills/drolu/agent-skills/excel/Gen Agent Trust Hub

excel

Pass

Audited by Gen Agent Trust Hub on Jul 2, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes a helper script, recalc.py, which executes LibreOffice (soffice) and system timeout utilities using the subprocess module. The implementation uses list-based argument passing, which prevents shell injection vulnerabilities.
  • [PROMPT_INJECTION]:
  • Ingestion points: The skill is designed to ingest and process untrusted external data from various spreadsheet formats (.xlsx, .xlsm, .csv, .tsv) as described in SKILL.md.
  • Boundary markers: The instructions do not define specific delimiters or "ignore instructions" warnings to wrap cell data when it is processed by the agent.
  • Capability inventory: The skill provides file system read/write access and system command execution capabilities through the included recalc.py script and the pandas/openpyxl libraries.
  • Sanitization: Current instructions focus on data integrity and formula verification rather than security-focused sanitization of spreadsheet content to prevent indirect prompt injection.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 2, 2026, 07:11 PM
Security Audit — agent-trust-hub — excel