Meridian wants a single assistant that can answer any question a terminal planner asks against the entire library of port-authority circulars, complete with citations, and the team has no interest in maintaining custom retrieval or prompt-engineering code for it. Which approach best fits this requirement?
Select an answer to reveal the explanation.
Short Explanation
This is the 'point it at a pile of documents and let it answer questions with citations, and please don't make me maintain the plumbing' request - that's exactly the job the Knowledge Assistant pattern was built to do out of the box.
Full Explanation
Agent Bricks Knowledge Assistant is designed for precisely this scenario: a single corpus, open-ended question answering over it, citation-backed responses, and minimal custom engineering, since it manages the retrieval and grounding machinery for the team. A hand-built custom RAG pipeline could technically achieve the same outcome, but the team explicitly said it does not want to build or maintain custom retrieval or prompt code, which rules out the DIY path even though it is technically capable. Multi-Agent Supervisor is the wrong pattern because there is only one knowledge domain and one type of question here - supervisor patterns exist to route work across multiple specialized sub-agents, and introducing one when a single knowledge base and a single answering behavior are all that's needed is unnecessary complexity. Information Extraction is the wrong pattern because the ask is open-ended Q&A over a document library, not pulling a fixed set of structured fields out of documents at scale. Recognizing when a managed Agent Bricks pattern is the right call versus over- or under-fitting the tool to the job is a core Section 1 skill.