p-image

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.70). The SKILL.md examples for Image Editing and Multi-Image Compositing explicitly accept arbitrary public image URLs (e.g., "images": ["https://your-image.jpg"], "images": ["https://img1.jpg", ...]) which the app will fetch and process, so the agent ingests untrusted third-party content that could carry instructions or influence outputs.