sap-api-policy

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

• Third-party content exposure detected (high risk: 0.75). The required workflow fetches evidence from web/official sources (notably the “Web search fallback” path, plus various MCP fetch/search tools that retrieve public text), so the LLM can ingest outsider-authored free text from third-party web pages/HTML at runtime when MCP sources aren’t available.