skills/carlheath/ogmios/pptx/Gen Agent Trust Hub

pptx

Pass

Audited by Gen Agent Trust Hub on Oct 7, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external .pptx files by extracting their content for analysis and editing, creating a potential surface for indirect prompt injection.
  • Ingestion points: Text is extracted from presentation.pptx using the markitdown tool as described in the Reading Content section of SKILL.md.
  • Boundary markers: The skill does not define specific delimiters or instructions to ignore potential commands embedded within the processed PowerPoint files.
  • Capability inventory: The skill utilizes shell commands to run local Python scripts (scripts/thumbnail.py, scripts/office/unpack.py) and system utilities (pdftoppm, soffice).
  • Sanitization: No explicit sanitization or filtering of the extracted text content is mentioned before it is analyzed by the agent.
  • [COMMAND_EXECUTION]: The skill relies on shell command execution for its core functionality.
  • Evidence: Instructions include running python scripts/thumbnail.py, python scripts/office/unpack.py, and pdftoppm for visual QA and file manipulation. These commands are scoped to the project's local directory and standard system tools.
  • [EXTERNAL_DOWNLOADS]: The skill documents the need for several external libraries.
  • Evidence: The skill references pip install "markitdown[pptx]", pip install Pillow, and npm install -g pptxgenjs. These are well-known packages from established providers and official registries.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 7, 2026, 02:14 AM