javascript-sdk

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly includes RAG/web-search patterns that call public search tools and enable webSearch (e.g., references/agent-patterns.md’s RAG Pattern using appTool('search','tavily/search-assistant@latest') and references/tool-builder.md showing internalTools().webSearch(true)), so the agent will fetch and ingest untrusted public web content and use it to influence responses and tool actions.