eachlabs-image-generation

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

• Third-party content exposure detected (high risk: 0.70). The skill's model parameter reference (references/MODELS.md) clearly allows user-supplied public image URLs and image_prompt_urls/image_reference_url (e.g., "image_url", "image_prompt_urls", "image_reference_url") which the EachLabs Predictions API will ingest and use to influence generation, so it consumes untrusted third‑party content that can materially affect outputs.