imagekitio

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

• Third-party content exposure detected (high risk: 0.70). The skill explicitly instructs the agent to call ImageKit.io via Membrane actions and proxy requests (e.g., "list-files", "get-file-details", "get-file-metadata" and "membrane request ...") which fetch user-uploaded, third-party media and metadata that the agent is expected to read and that can drive follow-up actions like purge-cache or transformations, so untrusted content could indirectly influence behavior.