media-understanding

Pass

Audited by Gen Agent Trust Hub on Apr 30, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements a secure wrapper for the Gemini API using the official google-genai Python library.
  • [SAFE]: Sensitive information, specifically the GEMINI_API_KEY, is managed through environment variables or local .env files, which is the recommended practice for secret management.
  • [SAFE]: The scripts include automated cleanup logic to delete files from the Google Gemini Files API once analysis is finished, minimizing the data footprint.
  • [PROMPT_INJECTION]: As the skill is designed to analyze arbitrary local media files (audio/video), it is inherently susceptible to indirect prompt injection if the processed media contains spoken or visual instructions designed to manipulate the AI. This is a standard risk for multimodal processing tools and does not indicate a flaw in the skill's code.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 30, 2026, 10:06 AM
Security Audit — agent-trust-hub — media-understanding