pptx
Pass
Audited by Gen Agent Trust Hub on Mar 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill relies on executing several local Python and JavaScript scripts provided by the author to handle OOXML unpacking, packing, and text replacement. It also utilizes system binaries such as 'soffice' (LibreOffice) for PDF conversion and 'pdftoppm' for image generation.
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests and processes text content from external .pptx files which could contain malicious instructions.
- Ingestion points: Presentation content extracted via 'markitdown', 'inventory.py', and raw XML access.
- Boundary markers: No specific boundary markers or 'ignore' instructions are provided to the agent when processing extracted text.
- Capability inventory: Full file system access for presentation files, subprocess execution for scripts and system tools, and headless browser rendering via Playwright.
- Sanitization: The skill explicitly recommends 'defusedxml' for secure XML parsing to prevent XXE attacks, but does not implement sanitization for natural language instructions found within document content.
- [EXTERNAL_DOWNLOADS]: The documentation includes instructions for installing standard dependencies from well-known registries (PyPI, NPM) and official repositories, which is typical for developer-oriented tools.
Audit Metadata