ml-dl-expert

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

• Third-party content exposure detected (high risk: 0.90). This skill explicitly instructs ingestion of external content — e.g., the /find-dataset flow ("public datasets") and the data-pipeline / rag-retrieval paths that parse PDFs, chunk documents and build RAG/vector stores (FAISS, ChromaDB) — meaning the agent will fetch and read untrusted third-party/user-provided content as part of its workflow.