cloudinary-transformations

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

• Third-party content exposure detected (medium risk: 0.30). SKILL.md defines a natural-language-to-Cloudinary URL generator and includes guidance for “fetch remote image” delivery URLs (i.e., URLs that instruct Cloudinary to retrieve external content), so attacker-authored free text can be incorporated into the transformation parameters/prompt-like fields used at runtime.