skills/kang-chen/agent-skills/pptx/Gen Agent Trust Hub

pptx

Pass

Audited by Gen Agent Trust Hub on Mar 1, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill parses content from user-provided .pptx files, creating a surface for indirect prompt injection. • Ingestion points: 'scripts/inventory.py' extracts all text from slides into a JSON inventory. • Boundary markers: The extracted text is structured in JSON paragraph objects but lacks explicit delimiters or instructions to ignore embedded commands. • Capability inventory: The skill can execute local commands, render HTML via a headless browser, and perform extensive file system operations. • Sanitization: XML content is processed using 'defusedxml' for all XML parsing to mitigate XML External Entity (XXE) vulnerabilities, which is a significant security control.
  • [COMMAND_EXECUTION]: Local utilities like LibreOffice and Poppler are used for file conversions and visual analysis. • Evidence: 'scripts/thumbnail.py' and 'ooxml/scripts/pack.py' invoke 'soffice' and 'pdftoppm' using 'subprocess.run'. • Context: These calls use argument lists rather than raw shell strings, preventing shell injection. These operations are necessary for the skill's primary functionality.
  • [EXTERNAL_DOWNLOADS]: The skill documentation requires standard development and productivity tools. • Evidence: Installation instructions for 'markitdown', 'pptxgenjs', 'playwright', and 'sharp' are provided in 'SKILL.md'. • Context: All dependencies are sourced from trusted official registries (npm and PyPI) or standard OS repositories (apt), which are recognized as safe sources.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 1, 2026, 08:45 PM