skills/revfactory/skills/hwp/Gen Agent Trust Hub

hwp

Pass

Audited by Gen Agent Trust Hub on Aug 15, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [SAFE]: The skill acts as a comprehensive technical guide for handling the HWP and HWPX document formats. The provided code for binary parsing and XML handling follows standard development practices.
  • [COMMAND_EXECUTION]: The documentation references legitimate command-line utilities such as pyhwp and LibreOffice for document conversion tasks, which is consistent with the skill's purpose.
  • [DATA_EXPOSURE]: The skill includes logic for reading and writing local files as part of document conversion workflows. It incorporates resource safety measures, such as skipping encrypted files and limiting the size and quantity of extracted images to prevent performance issues.
  • [INDIRECT_PROMPT_INJECTION]: The skill defines a surface for processing external document data.
  • Ingestion points: Files are read from the local file system using fs.readFileSync in Node.js and olefile or zipfile in Python.
  • Boundary markers: The instructions focus on structural parsing and do not specify distinct boundary markers for the extracted text content.
  • Capability inventory: The skill enables file reading, text extraction, image extraction, and file generation.
  • Sanitization: The skill provides an escapeXml function to properly handle special characters during the generation of HWPX documents.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 15, 2026, 08:40 AM
Security Audit — agent-trust-hub — hwp