ai-multimodal

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 analyzes public, user-generated content (e.g., YouTube videos and arbitrary public image/video URLs) — see references/video-analysis.md (YouTube example using types.Part.from_uri) and vision-understanding.md (URL example) and the batch scripts that pass those URIs to client.models.generate_content — so the agent will fetch and interpret untrusted third-party content.