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

# Introduction

> Safe, governed database access for AI agents.

TeamDuo lets a team expose selected parts of its databases to AI agents — read-only,
scoped to the tables and columns you choose, and annotated with the context an agent
needs to query them correctly.

<CardGroup cols={2}>
  <Card title="Quickstart" icon="rocket" href="/quickstart">
    Connect your first datasource and run a query in a few minutes.
  </Card>

  <Card title="Core concepts" icon="book-open" href="/concepts">
    Teams, datasources, exposure rules, and how queries are checked.
  </Card>

  <Card title="Connect a datasource" icon="database" href="/guides/connect-a-datasource">
    Point TeamDuo at Postgres and choose what to expose.
  </Card>

  <Card title="MCP tools" icon="plug" href="/reference/mcp-tools">
    The tools agents call, and what each one returns.
  </Card>
</CardGroup>

## How it works

1. **Connect** a database and grant TeamDuo read-only credentials.
2. **Expose** the tables and columns your team is comfortable sharing, with notes
   describing what each one means.
3. **Query** from any MCP-compatible client. Anything outside the exposed surface is
   refused with an explanation rather than silently returning data.

<Note>
  This documentation is published from the
  [`devduoai/teamduo-docs`](https://github.com/devduoai/teamduo-docs) repository.
  Edits merged to `main` deploy automatically.
</Note>