search-specialist

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly performs web research and "WebFetch Deep Dive" to "extract full content from promising results" and returns "curated findings with source URLs" — i.e., it fetches and ingests content from public web pages, which can be untrusted/user-generated and thus could carry indirect prompt injections.