cloudlayer

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly describes actions that fetch and ingest arbitrary public webpages (e.g., "URL-to-image", "URL-to-pdf", and "extract website content" in the Cloudlayer overview and Popular actions), so the agent may read untrusted third-party page content that could influence subsequent decisions or tool use.