> ## Documentation Index
> Fetch the complete documentation index at: https://docs.getvirtualbrain.com/llms.txt
> Use this file to discover all available pages before exploring further.
# Overview
> Ask in plain language, get answers traced to their source, and turn them into finished deliverables, end to end
Chat is where most of your day-to-day work happens: you select one or more [Brains](/features/brains/overview), ask a question in plain language, and VirtualBrain answers using those documents (plus the web when you switch it on), citing every source so you can verify each claim and turn the answer into a finished deliverable.
## Why Chat is built around verification
In advisory work, every claim you put in front of a client carries your firm's name. Checking those claims costs more than generating them, which is where the productivity gain usually disappears.
Chat is a workflow built for human review: one-click citations, side-by-side verification, fully editable output, and targeted regeneration of just the paragraph you don't like. The AI drafts; you verify and decide.
## How a chat unfolds
Every chat follows the same arc: ground your question in the right sources, watch the [Mainframe](/features/mainframe) research and assemble a sourced answer, then shape the deliverable and export it. Each stage has its own actions:
Start a chat, choose the Brains and files that ground the answer, and prompt for precision.
Watch the research happen, then verify every claim against its source, side by side.
Revise the deliverable in place, then export it to Word, PDF, Excel, or PowerPoint.
Behind all three sits the same engine.
What it can reach in a chat: your documents, the web, your tables analyzed, and a comparison report between two document versions
## How a chat unfolds
Every chat follows the same arc: ground your question in the right sources, watch the [Mainframe](/features/mainframe) research and assemble a sourced answer, then shape the deliverable and export it. Each stage has its own actions: