pptx
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes command-line tools such as LibreOffice (soffice), Poppler (pdftoppm), and git for presentation conversion and comparison. These operations are performed using argument lists in subprocess calls, which mitigates shell injection risks. Findings located in scripts/thumbnail.py, ooxml/scripts/pack.py, and ooxml/scripts/validation/redlining.py.\n- [DYNAMIC_EXECUTION]: The html2pptx.js script uses the Playwright library to render HTML slides in a headless browser environment. This enables the calculation of precise element positioning for PowerPoint generation. The execution is confined to layout rendering from local files.\n- [INDIRECT_PROMPT_INJECTION]: The skill extracts text content from user presentations, which creates a vulnerability surface for indirect prompt injection.\n
- Ingestion points: Slide text extraction in markitdown and inventory.py.\n
- Boundary markers: Absent; the agent instructions do not mandate delimiters when reading extracted slide content.\n
- Capability inventory: Shell command execution, file system access, and headless browser rendering.\n
- Sanitization: Absent; the extracted text is processed without explicit filtering.
Audit Metadata