infographic-generation

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

• Third-party content exposure detected (high risk: 0.80). The skill's SSE event docs (references/SSE-EVENTS.md) define and demonstrate runtime web_search_query and web_search_citations events that fetch and ingest open-web URLs, which the agent uses to inform responses (e.g., "Based on current information..."), so untrusted third-party content can materially influence its actions.