primary-logic

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

• Third-party content exposure detected (high risk: 1.00). The SKILL.md explicitly ingests open/public third‑party content (e.g., podcasts, articles, X/Twitter, Kalshi, Polymarket) via required MCP calls like search_content and get_content/get_ticker_content, which the agent is instructed to read and use to drive signal-ranked decisions, so untrusted user-generated content could indirectly inject instructions.