pptx

Pass

Audited by Gen Agent Trust Hub on Jun 28, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection.
  • Ingestion points: It processes user-provided .pptx files in the 'Reading Content' and 'QA (Required)' sections of SKILL.md.
  • Boundary markers: No delimiters or specific 'ignore embedded instructions' warnings are present in the visual QA prompt templates used for subagent analysis.
  • Capability inventory: The skill utilizes subprocess calls for file conversion (python, pdftoppm, soffice) and package management (npm, pip) in SKILL.md.
  • Sanitization: There is no evidence of sanitization or content validation performed on the extracted slide text or visual elements before they are evaluated by the AI.
  • [COMMAND_EXECUTION]: The skill executes several command-line tools to manipulate and verify files.
  • Operations include text extraction via 'markitdown', format conversion via 'soffice', and image generation via 'pdftoppm'.
  • These commands are run against user-supplied filenames, which requires the agent to ensure proper shell escaping.
  • [EXTERNAL_DOWNLOADS]: The skill installs necessary dependencies from public registries.
  • It fetches 'markitdown' and 'Pillow' from PyPI, and 'pptxgenjs' via NPM. These are well-known and standard libraries for presentation and image processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 28, 2026, 01:28 PM
Security Audit — agent-trust-hub — pptx