docx

Pass

Audited by Gen Agent Trust Hub on Mar 6, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes standard system utilities including pandoc for text extraction, unzip/zip for document manipulation, and soffice (LibreOffice) for PDF conversion. These are established tools for document processing workflows.
  • [EXTERNAL_DOWNLOADS]: The skill references the docx library for Node.js, which is a well-known and reputable package for programmatic document creation.
  • [PROMPT_INJECTION]: The skill processes untrusted external .docx files, creating an ingestion surface for potentially malicious instructions. 1. Ingestion points: External document files processed via pandoc, unzip, and soffice in SKILL.md. 2. Boundary markers: Absent. 3. Capability inventory: Execution of system commands (pandoc, unzip, zip, soffice, pdftoppm) and file system writes (fs.writeFileSync) throughout SKILL.md. 4. Sanitization: No explicit sanitization or instruction-filtering is mentioned for the extracted text.
  • [SAFE]: No signs of data exfiltration, obfuscation, or malicious intent were detected. The skill is provided by a trusted source and its operations are necessary for document management.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 6, 2026, 05:29 PM