blogpost-figure

Pass

Audited by Gen Agent Trust Hub on Oct 2, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The script scripts/render_html.py uses subprocess.run to invoke a headless browser (Google Chrome or Chromium) to convert HTML templates into PNG images. This is the primary function of the skill and the commands are constructed as lists for secure argument passing.
  • [EXTERNAL_DOWNLOADS]: The scripts/fetch_fonts.py script downloads open-licensed font files from fonts.googleapis.com. This is a well-known and trusted service used to maintain consistent typography in the generated figures.
  • [DYNAMIC_EXECUTION]: The scripts/render_html.py script performs minor runtime pre-processing of HTML files to resolve local asset paths, which is a standard requirement for rendering complex HTML templates.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied data to populate chart labels, titles, and card content. While this creates a surface for indirect prompt injection, it is the intended use-case and is managed by the agent's safety protocols.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 2, 2026, 08:53 PM