docx

Warn

Audited by Gen Agent Trust Hub on Mar 21, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute system-level commands, such as pandoc for text extraction, soffice for PDF conversion, and pdftoppm for image processing.
  • [COMMAND_EXECUTION]: Explicit instructions are provided to use sudo apt-get install for system dependency setup, which involves acquiring elevated administrative privileges.
  • [COMMAND_EXECUTION]: The primary workflow relies on the agent generating and running custom Python and JavaScript scripts at runtime to manipulate document structures.
  • [EXTERNAL_DOWNLOADS]: The skill fetches dependencies from standard public registries, such as docx via NPM and defusedxml via Pip. It also references external XML schema definitions from dublincore.org.
  • [PROMPT_INJECTION]: Processing external documents (.docx) creates a surface for indirect prompt injection, where instructions embedded in document text could be interpreted by the agent. The skill integrates defusedxml to provide protection against common XML vulnerabilities.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 21, 2026, 02:43 AM
Security Audit — agent-trust-hub — docx