You must integrate enterprise knowledge for multi-agent consumption: product manuals in search indexes, policy packs via MCP-available sources, and semantic retrieval for fuzzy engineer questions. Which design best implements knowledge integration?
Select an answer to reveal the explanation.
Short Explanation
You need a knowledge plane that lets agents retrieve authorized content through consistent tools. Combining search, RAG, MCP-available sources, and semantic search keeps specialists grounded without leaking data or relying on stale model memory.
Full Explanation
The correct design combines search, RAG, MCP-available sources, and semantic search so agents retrieve authorized content through consistent tool interfaces. Multi-agent development needs a governed knowledge plane: retrieval, drafting, and validation agents share grounded content when retrieval tools enforce tenant and access-control context. Relying solely on the base model cut-off is wrong because base models lack private enterprise updates. Uploading secrets and private customer data into a public demo chatbot is wrong because it violates security. Emailing PDFs to an orchestrator mailbox and parsing them only with regex is wrong because it is not robust or scalable. Enforce access control at retrieval time, not only in prompts.