flux-image

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

• Third-party content exposure detected (high risk: 0.70). The skill's "Edit Images" workflow explicitly accepts an arbitrary image_url (e.g., "https://example.com/photo.jpg") which causes the API to fetch and interpret untrusted third-party image content as part of its editing workflow, allowing that content to materially influence model outputs.