pptx
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local shell commands and scripts to handle file conversions and extraction tasks.
- Evidence: Executes
soffice.py(LibreOffice) andpdftoppm(Poppler) for document conversion. - Evidence: Invokes
markitdownfor text extraction from PowerPoint files. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted .pptx files, creating a potential surface for instructions embedded in document content to influence agent behavior.
- Ingestion points: .pptx file content processed in SKILL.md via markitdown.
- Boundary markers: None identified in instructions to delimit extracted text.
- Capability inventory: Subprocess execution (soffice, pdftoppm), file system writes, and package installation.
- Sanitization: Not explicitly implemented for extracted document text.
Audit Metadata