The assistant experience

This is what your users actually see when they open your assistant — a full, modern chat, entirely under your brand. It works like the AI chat apps your users already know, so there's nothing new to learn.

The chat, end to end

• Streaming, formatted answers. Responses stream in live and render as rich text — headings, lists, tables, and code — not a wall of plain text.
• Conversation history. Every chat is saved. Your users switch between conversations from a sidebar, reopen any one right where they left off, and each thread is auto-titled from what they asked. They can rename or delete their own conversations.
• Projects. Related conversations can be grouped into projects in the side rail, so a user working on one topic keeps those chats together.
• A model picker. Users can pick which model answers — subject to your policy. If you pin a single model for your tenant, the assistant stays on it. See Governance & model policy.
• A path to a human. When the assistant can't finish something, your user can ask for a person, and a member of your team steps into the same conversation. See Human support & handoff.

Signed in, or as a guest

How much a user sees depends on how you set up identity (see Getting started):

• Signed-in users get their own private history and — if you've connected your tools — actions taken on their account, scoped to just their data.
• Guests (if you enable guest access) can still chat and get answers and guidance, without signing in. Guests don't get actions on a specific user's private data.

It's your brand, everywhere

The name at the top, the colors, the welcome message, the avatar — all yours. The same assistant can run as a full page on your host and as an embedded panel inside your product, and it looks and behaves the same in both. Your users never see our brand; they see you.

Consistent light and dark. The assistant renders correctly in both light and dark themes automatically, following your users' system preference.

Next

• Governance & model policy — decide which features are on and how models are chosen.
• Human support & handoff — how a person joins the chat.
• Connecting your tools — let the assistant act, not just answer.