gemini-media-gen

Pass

Audited by Gen Agent Trust Hub on Jul 10, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements a mandatory 'Confirm before sending' step using the AskUserQuestion tool. This ensures the user reviews the expanded engine brief, model selection, and destination path before any external API calls or script executions occur.
  • [SAFE]: External dependencies are limited to the official google-genai Python library, a standard package from a well-known service provider (Google).
  • [SAFE]: The skill correctly manages sensitive data by instructing users to provide the GEMINI_API_KEY via environment variables rather than hardcoding it, following industry best practices for credential management.
  • [SAFE]: Network activity is restricted to communication with Google's Generative AI API endpoints via the official SDK for the purpose of media generation and status polling.
  • [SAFE]: File system access is transparent; the skill reads user-provided reference images and saves outputs to a location explicitly confirmed by the user, defaulting to a project-specific folder or the Desktop.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 10, 2026, 06:01 AM
Security Audit — agent-trust-hub — gemini-media-gen