gemini-video-understanding
Warn
Audited by Snyk on Mar 1, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.90). The skill explicitly accepts and processes public YouTube URLs (see SKILL.md, EXAMPLES.md and scripts/analyze_video.py where --youtube-url is passed as a file_uri to the Gemini API), so the agent ingests untrusted, user-generated third‑party content that can directly influence model outputs and downstream actions.
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 0.80). The skill accepts and sends YouTube URLs (e.g. https://www.youtube.com/watch?v=VIDEO_ID) at runtime directly as a file_uri to the Gemini API, causing remote video content to be fetched and injected into the model's context (i.e., directly controlling the agent's inputs/outputs).
Audit Metadata