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

# Database

> Choose the knowledge sources (RAG) this agent will use to enrich its answers.

From an agent's **Database** tab, select the knowledge sources this agent can draw on to build its answers, through Retrieval Augmented Generation (RAG).

<Frame>
  <img src="https://mintcdn.com/weengage/Gl3GO9m2EAsc7X0J/images/agents/general-information/volubile-agent-rag.png?fit=max&auto=format&n=Gl3GO9m2EAsc7X0J&q=85&s=4c4efc0353f54e32987339b326565b0a" alt="Selecting the agent's data sources" width="1372" height="884" data-path="images/agents/general-information/volubile-agent-rag.png" />
</Frame>

## Selecting sources

The page mirrors your entire knowledge base — **Documents** and **Web pages** — with a checkbox in front of each source.

| Element                | Description                                               |
| ---------------------- | --------------------------------------------------------- |
| **Tick a source**      | The source becomes available to this agent.               |
| **Multiple selection** | You can enable as many documents and web pages as needed. |
| **No box ticked**      | No source is attached to the agent: it will not use RAG.  |

<Note>
  The selection is **specific to each agent**. It does not change the base content or other agents. Remember to save
  the agent configuration to apply your changes.
</Note>

<Accordion title="What is RAG?">
  Retrieval Augmented Generation (RAG) is a technique that allows your agent to access specific information stored in an external knowledge base to enrich its responses with accurate and contextual data.
</Accordion>

<Tip>
  Importing documents, adding web pages and editing fragments are managed from the
  [Database](/en/product/rag) page, shared across the customer.
</Tip>