Knowledge Base

So it answers with
your material, not its guesses.

Upload your manuals, your return policy, your catalog and your FAQs. Before answering, the agent searches there. If it is not there, it says so.

What you can load

What you already have written down somewhere.

  • Files

    PDF, Word, spreadsheets and text. Processed and available in minutes.

  • Websites

    Give it an address and it reads it. It refreshes itself periodically.

  • What's in your Drive

    Files you already uploaded to Napsix can be added without uploading them again.

How an agent uses it

It searches before it speaks, and tells you where it got it.

Not a giant prompt: a real search over your documents, scoped to what that agent is allowed to see.

  • Each agent only reaches what you assigned to it
  • The answer cites the document it came from
  • If the answer is not there, it says so instead of improvising
  • A project can have its own base, separate from the rest
FAQ

What people usually ask.

Do my documents train any model?
No. They are used to answer your queries and nothing else. The providers we use operate with zero retention and zero training on your data.
What if I update a document?
Upload it again and it replaces the previous one. Web sources refresh themselves.
Start today

Upload a document and ask it something.

Free to start. It answers from it within minutes.

XIA
public