industry-chain-intelligence

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

• Third-party content exposure detected (medium risk: 0.30). Required runtime Phase 1 performs “并行数据采集” by calling multiple first-order MCP tools (e.g., industry_tech_chain, company_word_cloud, etc.) to fetch database-derived text used in Phase 2, and while the outsider provides only the company/chain string (not arbitrary content), the MCP-returned fields are runtime-ingested free text that could include outsider-authored strings if the underlying data source contains user-submitted content.