paw-tools-presentation

Pass

Audited by Gen Agent Trust Hub on Apr 2, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches libraries from well-known services, including Chart.js via jsDelivr and Tailwind CSS via tailwindcss.com. These are used to provide the presentation with interactive data visualizations and styling.
  • [EXTERNAL_DOWNLOADS]: HTML templates in the assets folder contain placeholder images hosted on Google (lh3.googleusercontent.com), which serve as professional portraits for team and contact slides.
  • [PROMPT_INJECTION]: The skill processes untrusted marketing content provided via raw text, file paths, or other skill outputs (Category 8 surface). This ingestion is central to the skill's primary function of content transformation. Ingestion point: references/01-ingest.md (reads raw text, file content, and previous skill outputs). Boundary markers: Absent. Capability inventory: File system write operations (references/03-generate.md). Sanitization: Absent for content placeholders. The risk is minimized by the agent's internal safety filters and the static nature of the generated HTML output.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 2, 2026, 04:35 AM