skills/jimliuxinghai/skills/docx/Gen Agent Trust Hub

docx

Pass

Audited by Gen Agent Trust Hub on Mar 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill strictly uses the defusedxml library for all XML manipulation in scripts/document.py and associated scripts, effectively mitigating risks of XML External Entity (XXE) and billion laughs attacks.
  • [SAFE]: Command execution is limited to necessary document processing tools (pandoc, soffice, pdftoppm) and is implemented without shell invocation vulnerabilities, using pathlib for secure path handling.
  • [SAFE]: The skill operates on untrusted document content by design. While this presents a surface for indirect prompt injection, the skill includes a dedicated validation suite (ooxml/scripts/validation/) that ensures document integrity and validates the structure of tracked changes against original versions.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 13, 2026, 01:50 AM