macro-event-analysis

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly instructs the agent to fetch and ingest event data from public third-party providers (see SKILL.md and references/providers/fmp.md and references/providers/tradingeconomics.md, which include curl calls to FMP and TradingEconomics) and to use that information to build the event slate and drive analysis, so untrusted external content can materially influence decisions.