map

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

• Third-party content exposure detected (medium risk: 0.30). [SKILL.md] describes a workflow that uses “搜索…并用 defuddle/WebFetch 提取正文” to collect and read outsider-authored materials from the user-selected web/search sources; since the runtime ingests the extracted article/document text without requiring selection beyond “search and fetch,” this is an indirect prompt-injection exposure path.