report-to-google-slides

Pass

Audited by Gen Agent Trust Hub on Jul 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • Structured Data Processing: The skill utilizes a Python script to parse HTML and generate a local PowerPoint (.pptx) file. This involves local file system operations and command execution that are necessary for the report conversion process.
  • Content Sanitization: The report_to_google_slides.py script includes a comprehensive sanitize_svg_markup function. It explicitly filters out potentially risky HTML and SVG tags such as script, iframe, object, and embed, and strips JavaScript event handlers (e.g., onclick). This mitigates risks associated with processing untrusted report content.
  • External Resource Management: The rendering logic for charts is designed to be self-contained. While it attempts to use available system tools like cairosvg or playwright for high-fidelity rendering, it defaults to localized rendering methods and implements a page.route("**/*", lambda route: route.abort()) policy during headless browsing to prevent the loading of external network resources.
  • Safe Dependency Usage: The skill references standard Python libraries such as beautifulsoup4, pillow, and python-pptx. These are well-established packages in the Python ecosystem for document and image processing.
  • Implicit Safety Guidelines: The SKILL.md file provides clear instructions to the agent to verify reports manually and stop if redirect or sign-in pages are encountered, reinforcing safe usage patterns.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 15, 2026, 09:17 AM
Security Audit — agent-trust-hub — report-to-google-slides