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

# Automated Onboarding

> Build your knowledge base automatically from your data warehouse

Delphina analyzes your warehouse — schemas, query history, expert users' SQL, and uploaded documents — and documents your most important tables, metrics, and business rules automatically.

## Prerequisites

* A data warehouse is connected. See [Warehouse Connections](/administration/warehouse-connections).
* You have the Developer role.

## Step 1: Configure sources

Tell Delphina where to look. Go to **Context > Sources** and configure the raw inputs the agent will use to build your knowledge base.

**Tables** — Select which warehouse tables should be in scope. Browse your schema and check the tables you want documented. Use **Bulk Edit** to paste table names if you have many.

**Query Users** — (Snowflake/BigQuery only) Select the users whose query history the agent should analyze. Pick your most experienced analysts — the agent learns join patterns, filter conventions, and metric definitions from their SQL.

**File Uploads** — Upload reference documents that describe your data: dashboard definitions (LookML, Tableau), dbt project files, data dictionaries, or any other documentation. The agent extracts metric definitions, business rules, and table relationships from these.

**MCP Connections** — Connect external tools and APIs via the Model Context Protocol for additional context. See [MCP Connections](/administration/mcp-connections).

## Step 2: Update knowledge

Click **Update Knowledge** at **Context > Jobs**. The agent processes all configured sources — schemas, query history, uploaded documents, and MCP connections — and creates table documentation, metric definitions with tested SQL, and business rules.

This may take some time for larger warehouses. Progress is visible on the jobs page.

## Step 3: Create evaluations

Click **Create Evals** at **Context > Evaluations**. The agent generates test cases from the knowledge it just built — questions paired with expected SQL and acceptance criteria.

Review the candidate evaluations with your team. These validate that the agent answers questions consistently with your documented definitions. See [Evaluations & Quality](/trust-and-quality/evaluations).

## After onboarding

The agent immediately starts using your documented metrics, rules, and tables. Evaluations run weekly to catch regressions.

Use `/knowledge` for ongoing refinement as schemas change. Re-run **Update Knowledge** when expanding to a new domain or after updating sources. See [Maintenance](/knowledge-agent/maintenance).