> ## Documentation Index
> Fetch the complete documentation index at: https://chatobserver.com/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Workspace Tour

> A practical overview of the main Chatobserver product surfaces.

# Workspace Tour

Chatobserver is organized around a small set of product surfaces that build on each other.

## Overview

The overview surface is where teams sanity-check the health of a workspace. It combines setup reminders, high-level brand visibility signals, and AI referrals metrics so you can see whether data is flowing and whether the top-level trends are moving in the right direction.

## Prompts

Prompts are the core unit of work in Chatobserver. Each prompt defines:

* the question to ask
* the platforms to run on
* the region to target
* the cadence for execution
* the grouping context for analysis

Everything downstream starts here.

## Runs

A prompt run is one execution of one prompt. Runs tell you whether a prompt executed successfully, when it started and finished, and whether the result has been analyzed.

If prompts are the plan, runs are the operational record.

## Sources

The sources surface turns raw AI answers into something a team can analyze. It focuses on domains, citations, mentions, and prompt-level source activity so you can understand which publishers and brands are appearing across your tracked prompt set.

## Reports

Reports are the presentation layer. They let you build structured views of workspace data and share read-only snapshots with stakeholders who do not need direct workspace access.

## Settings

Settings covers the admin side of the workspace:

* billing and usage
* team and access
* company and workspace preferences
* API keys and developer access

## Suggested working rhythm

For most teams, the practical loop looks like this:

1. Refine prompts.
2. Watch new runs land.
3. Review the sources that appear.
4. Turn the relevant slice into a report.
5. Adjust prompt coverage and repeat.