AI Influencer Generation

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

• Third-party content exposure detected (high risk: 0.80). The skill's SKILL.md explicitly accepts external reference images via image_urls (e.g., the "Using Reference Images" section and multiple curl examples with https://your-storage.com/... and https://example.com/...), which the each::sense API ingests to guide generations, so untrusted third-party content can directly influence model behavior.