frontend-slides
Pass
Audited by Gen Agent Trust Hub on Mar 18, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted text from external PowerPoint files, creating a vulnerability to indirect prompt injection where instructions within the data could influence agent behavior.
- Ingestion points: PowerPoint (.pptx) files processed in Phase 4.
- Boundary markers: None; extracted text is used directly in presentation generation.
- Capability inventory: Python script execution, file writing, and HTML/JS generation.
- Sanitization: None; text is extracted and processed without validation or filtering.
- [COMMAND_EXECUTION]: The skill provides Python code to be executed by the agent for extracting PPTX content. This code performs file system operations, including creating directories and writing image files to the local disk.
Audit Metadata