Skills
skills/kjgarza/marketplace-claude/pptx/Gen Agent Trust Hub

pptx

Pass

Audited by Gen Agent Trust Hub on Mar 6, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes system commands to facilitate presentation processing.\n
  • ooxml/scripts/pack.py and scripts/thumbnail.py utilize subprocess.run to call soffice (LibreOffice) and pdftoppm (Poppler) for document validation and image generation.\n
  • scripts/html2pptx.js employs Playwright to run a headless browser for rendering slide templates.\n- [PROMPT_INJECTION]: The skill possesses an indirect prompt injection attack surface because it processes content from external PowerPoint files.\n
  • Ingestion points: Text extraction occurs via markitdown and inventory.py.\n
  • Boundary markers: Not present; extracted text is provided to the agent without specific delimiters or warnings.\n
  • Capability inventory: The skill has the ability to execute system commands (subprocess.run) and launch a browser (Playwright).\n
  • Sanitization: No specific text sanitization is performed, though defusedxml is used for XML parsing.\n- [SAFE]: Security-conscious coding is evident in the usage of defusedxml throughout the XML handling scripts to protect against XXE attacks.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 6, 2026, 01:46 AM