qdrant

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

• Third-party content exposure detected (high risk: 0.90). The skill accepts arbitrary user/third-party documents via the /api/rag/ingest endpoint and stores them in Qdrant (EmbeddingStore), which the EmbeddingStoreContentRetriever and AiServices assistant then read and use for RAG—allowing untrusted content to materially influence model behavior.