longbridge-company-profile

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

• Third-party content exposure detected (high risk: 0.80). The skill's Workflow step 2 and CLI examples show it fetches "Latest news / catalysts (top 5)" and other market data via the longbridge news (and MCP fallback) commands, meaning the agent ingests third‑party public news/content which it uses to synthesize narratives and recommendations, creating a clear path for indirect prompt injection.