Welcome!

AI Agent per Library

TL;DR; Every library has its own AI agent — a chat experience that answers questions from your knowledge base's content: from one library when you ask from that library, or from all of its libraries that aren't archived with Ask the knowledge base. It's one of Lupo's most distinctive features, and there's nothing to turn on: anyone who can use the Knowledge Base chatbot opens it from the library.

Courses are good for sequenced learning. A browsable knowledge base is good for on-demand reference. An AI agent on top of that knowledge base is a third thing: a conversation that gives you a direct answer to a specific question, grounded in your organization's content.

What the AI agent is

The AI agent is a chat interface on your knowledge base. A learner types a question — "how long is parental leave?", "what's our incident response process?", "which tooling do we use for code reviews?" — and the agent answers using content items as its source material: the ones in a single library when they ask from that library, or the ones in every library that isn't archived when they use Ask the knowledge base.

A few things that make it specifically useful:

  • It only uses your knowledge base's content. The agent doesn't go to the broader internet, and when you ask from a library, it doesn't pull from the knowledge base's other libraries. Its answers are grounded in the content items you've uploaded.
  • It cites what it reads. Answers typically reference the specific content items they came from, so the learner can click through and verify.
  • Access is per knowledge base, not per library. Anyone who can use the agent on a knowledge base can ask about every library in it that isn't archived — see Library Visibility.

When it's most useful

The agent is there on every library, but it pays off most when:

  • The library has enough content to be useful. An agent pointed at three PDFs isn't much better than reading the three PDFs directly. The value kicks in once there are dozens of documents and the learner doesn't know where to start.
  • Learners ask the same questions repeatedly. If your HR team or Support team answers "how does PTO work?" five times a week, that's exactly what the agent is for.
  • The content is reasonably well-written. The agent answers from the text inside each file, so clear, well-structured documents get better answers. See Uploading Content Items for the file types it reads.

Don't point people at the agent on a library that's still being built out. Get the content in place first, then let them know it's there.

Opening the agent

There's no setting to turn on and nothing to configure. KB Members, KB Managers, and Company Administrators can open the agent from:

  • Ask this library inside a library, or Ask on the library's card in the list of libraries — answers from that library only.
  • Ask the knowledge base above the list of libraries — answers from every library that isn't archived.

The buttons don't appear on an archived library, and an archived library's content is left out of Ask the knowledge base answers. The chat page the buttons open is labelled Ask Lupo in its breadcrumb and browser tab, the same name as the chat in a course, but it answers from the knowledge base, not from a course.

Being enrolled in a course doesn't give access to the agent. A learner needs the KB Member role on the knowledge base.

Before you roll it out

A few things are worth knowing:

  • It's not a replacement for course-level tracking. The agent helps people find answers; it doesn't track whether they learned anything. For compliance training or formal onboarding, still use a course with activities and progress tracking.
  • It respects the data it's given. If the library has wrong or outdated information, the agent will confidently repeat it. Keep the content current — the AI makes bad content more visible, not less.

Tips for getting good answers

The biggest lever is the quality of your content, but there are a few other things that help:

  • Break long documents into smaller, focused files. A 200-page handbook with 40 topics is harder for the agent to answer from than 40 separate 5-page documents.
  • Use consistent terminology. If half your docs say "PTO" and half say "vacation days," the agent can still handle it, but learner questions line up with your content better when the vocabulary matches.
  • Review the agent's answers yourself. Before telling the whole department about it, spend 20 minutes asking it the questions you know learners will ask. You'll find gaps in the content that are easy to fix.

What the agent doesn't do

A few deliberate limitations worth knowing about:

  • It doesn't perform actions — it won't update documents, file tickets, or send emails.
  • It doesn't carry one chat into another — each chat draws only on its own messages. Your earlier chats are saved, though, and you can reopen them from the chat page.
  • It doesn't promise certainty. If a question has no good answer in the library, the agent says so rather than guessing.

If you need the agent to do any of those things, you're outside what this feature is for today.

Where to go next