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

# Connectors overview

> Connect your data sources to Interact so your agents can query real data.

## What are connectors?

Connectors give your agents access to your data. Each connector represents a secure, authenticated link between Interact and an external data warehouse. Once connected, agents can query that data in response to questions or scheduled jobs.

Without at least one connector, your agents can only work with information you've typed into their context — they can't retrieve live data.

## Getting started

<Steps>
  <Step title="Go to Settings → Connectors">
    In the left sidebar, navigate to **Settings** and click **Connectors**.
  </Step>

  <Step title="Click Add connector">
    Choose your data source from the list of supported systems.
  </Step>

  <Step title="Follow the setup guide for your system">
    Each connector type has its own setup guide with step-by-step instructions.
  </Step>

  <Step title="Attach the connector to an agent">
    When building or editing an agent, add a **Database Query Agent** subagent and select the connector from the dropdown.
  </Step>
</Steps>

## Supported systems

<Card title="Google BigQuery" icon="database" href="/org-settings/connectors/google-big-query">
  Connect your BigQuery project using a service account. Supports all BigQuery regions, standard SQL, and dataset-level access control.
</Card>

<Note>
  More data sources are on the roadmap. Reach out to your Turntwo account manager if you have a specific request.
</Note>

## Managing your connectors

From **Settings → Connectors**, you can:

* **View** all connectors in your organisation
* **Edit** an existing connector (update credentials or change dataset access)
* **Delete** a connector — note that any agent currently using it will lose access to that data

## One connector, multiple agents

A single connector can be used by multiple agents simultaneously. For example, you might have one BigQuery connector for your production data warehouse, shared across a Campaign Performance agent, a Budget Tracking agent, and an Anomaly Detection agent.

## Security

<AccordionGroup>
  <Accordion title="Encrypted credentials" icon="lock">
    Your service account JSON is encrypted using **Google Cloud KMS** before it's stored in Interact's database. Credentials are decrypted only at query execution time and are never exposed elsewhere.
  </Accordion>

  <Accordion title="Dataset whitelisting" icon="table">
    When you create a connector, you choose exactly which datasets (and optionally which tables) agents are permitted to access. Agents cannot query anything outside that scope — even if they try.
  </Accordion>

  <Accordion title="Query cost limits" icon="gauge">
    You can set a maximum query cost per connector (in BigQuery billing units). Interact estimates query costs before running them and blocks any query that would exceed your limit, protecting you from unexpected BigQuery bills.
  </Accordion>

  <Accordion title="Read-only access" icon="eye">
    Connectors are always read-only. Interact only runs `SELECT` queries — it never writes to, modifies, or deletes data in your warehouse.
  </Accordion>
</AccordionGroup>

[Deep dive: how connectors work →](/concepts/connectors)