skills/nickcrew/claude-cortex/pptx/Gen Agent Trust Hub

pptx

Pass

Audited by Gen Agent Trust Hub on Jul 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: Utility scripts such as ooxml/scripts/pack.py, scripts/thumbnail.py, and ooxml/scripts/validation/redlining.py execute system commands including soffice, pdftoppm, and git via subprocess.run. These operations are used for document conversion, validation, and diffing.
  • [COMMAND_EXECUTION]: The scripts/html2pptx.js script launches a headless browser using playwright to render local HTML slides. This is necessary for calculating element positions and CSS styles during the conversion process.
  • [PROMPT_INJECTION]: The skill handles untrusted data from PowerPoint files, creating a surface for indirect prompt injection. Ingestion points: scripts/inventory.py and the markitdown workflow in SKILL.md. Boundary markers: Content is read directly into context without specific delimiters. Capability inventory: The skill can write to the file system via scripts/replace.py and execute subprocesses. Sanitization: The skill employs defusedxml in ooxml/scripts/unpack.py and ooxml/scripts/pack.py to mitigate XML-based vulnerabilities like XXE.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 17, 2026, 04:47 PM
Security Audit — agent-trust-hub — pptx