multi-search-engine

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

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime workflow uses web_fetch to fetch public search result pages from outsider-operated sites (e.g., Google/DuckDuckGo/Brave/etc.) and then ingests the fetched page text into the agent’s LLM context, which is indirect prompt-injection exposure from arbitrary web content.