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

# Concepts

Supaboard connects your data, visualises it in dashboards, and lets you query it in plain English through AI agents.

# How everything fits together

```text theme={null}
Workspace
├── Data Sources   — your database / SaaS connections
├── Dashboard Groups
│   └── Dashboards — grids of widgets (charts, tables, KPIs, filters…)
├── Agents         — AI assistants trained on your data
├── Ask Stella     — chat interface powered by agents
└── Tools          — external integrations agents can call
```

# The 5 things you need to understand

## 1. Workspace

Your organisation's private environment. Everything — data, dashboards, teammates, agents — lives inside it. You can belong to multiple workspaces and switch between them from the sidebar.

## 2. Data Sources

Connections to your databases (PostgreSQL, BigQuery, Snowflake, MySQL, Sheets, and more). Widgets and Agents pull live data through these. Connect one before building anything else.

## 3. Dashboards & Widgets

A **Dashboard** is a drag-and-drop canvas of **Widgets**. Toggle between **Edit mode** (build) and **View mode** (consume). Widgets come in 9 types:

| Widget     | What it does                                                |
| ---------- | ----------------------------------------------------------- |
| **Chart**  | Bar, line, pie, scatter — any visualisation                 |
| **KPI**    | A single headline metric with trend                         |
| **Table**  | Paginated, sortable data grid                               |
| **Filter** | Viewer-interactive controls that update all widgets at once |
| **Text**   | Rich text block                                             |
| **Image**  | Image by URL                                                |
| **Embed**  | Any external URL in an iframe                               |

## 4. Agents

An AI assistant scoped to your data. You configure *what databases it can see*, give it a *knowledge base* and *verified queries*, and it learns to answer questions accurately over time. Agents power Ask Stella.

## 5. Ask Stella

A chat interface where you type a question in plain English — "What were our top 5 products last quarter?" — and Stella queries your data and answers. No SQL required.

# Other things worth knowing

* **Dashboard Groups** are folders that organise related dashboards. Sharing is set at the group level.
* **Roles** — Viewer → Editor → Admin → Owner. Admins can also restrict access per-feature or per-resource.
* **Export** — Download any dashboard or widget as a PDF or PowerPoint from the toolbar.
* **Mini-Board** — Embed dashboards in your own product with custom branding and feature toggles.
* **Tools** — MCP servers that extend what Agents can do (e.g. calling an internal API).

> **Start here:** Connect a Data Source → Create a Dashboard Group → Add a Dashboard → Drop in a Chart widget → Ask Stella a question about the same data.