A research agent actively goes out to gather new information from the web on a schedule. A RAG system (retrieval-augmented generation) answers questions by searching through a fixed library of documents you’ve already loaded. One is a scout; the other is a librarian.
Two Different Jobs, Two Different Tools
RAG stands for retrieval-augmented generation — a technique where an AI is given access to a specific library of documents and uses that library to answer questions rather than relying solely on its training data. If you’ve built a knowledge base of your course materials, community Q&As, and reference documents, a RAG system lets students or you query that library in natural language and get answers grounded in your actual content.
A research agent does something different. It doesn’t answer questions from a fixed library — it goes out to the live web, retrieves current information from the sources you’ve configured, and synthesizes what it finds. The key word is current. A RAG system knows everything in its library as of the last time you updated it. A research agent knows what happened yesterday.
When to Use Each
Use a RAG system when you want AI to answer questions based on your existing content — your course materials, past Q&A sessions, documentation, or knowledge base. This is powerful for student support (letting AI answer common questions by citing your actual course content) and for your own reference (quickly finding what you covered in a specific lesson without hunting through video recordings).
Use a research agent when you want ongoing awareness of what’s happening externally — in your niche, with competitors, in AI news, or in your community. The research agent’s job is discovery and monitoring. The RAG system’s job is retrieval and synthesis from a known set of materials. Both are valuable; they solve different problems.
What This Means for Educators
Many sophisticated AI-powered campuses use both: a RAG system connected to the course content library for student-facing Q&A, and a research agent running in the background to keep the educator informed about the external landscape. BetterDocs on a WordPress campus, for example, functions as a searchable knowledge base that a RAG system can draw from. The morning intelligence report delivered by a research agent feeds the educator’s real-time awareness. They’re complementary, not competitive.
The Simple Rule
If you’re asking “what did I teach about X?” — use a RAG system pointing at your content. If you’re asking “what’s happening with X in my industry right now?” — use a research agent. The scout goes out to find things; the librarian finds things you already have. Know which job you need done before deciding which tool to build.