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

# Agent Builder

> Turn a prompt into a working data extraction agent.

**Agent Builder** turns a natural language prompt into a working extraction Agent. Describe the site, fields, and record shape you want; Agent Builder plans the browse path, extracts sample data, validates the result, and saves a runnable Agent in your Space.

Use it when you know the data you want but do not want to hand-author commands. Reach for [the Editor](/agents/overview) when you need custom scripts, unusual navigation, or migration-level control.

<Note>
  Agent Builder is currently **single-shot**. Put the full request in your first prompt: URL, fields, pagination, authentication needs, output shape, and anything to skip.
</Note>

## How it works

<Steps>
  <Step title="Describe the target">
    Provide the URL, the fields to collect, and the shape of one record.
  </Step>

  <Step title="Builder creates the Agent">
    Agent Builder plans the navigation, browses the site, extracts records, and validates the output.
  </Step>

  <Step title="Run it anywhere">
    Open the saved Agent in Control Center, attach a [Schedule](/concepts/schedules), or [start Runs from the API](/api-reference/agents/start-run).
  </Step>
</Steps>

## Start building

<CardGroup cols={2}>
  <Card title="Get started with Agent Builder" icon="wand-magic-sparkles" href="/agent-builder/quickstart">
    Build your first Agent from Control Center with example prompts and expected output.
  </Card>

  <Card title="Use the Agent Builder API" icon="terminal" href="/api-reference/agent-builder/start-session">
    Start a Builder session from code, then poll status until it returns the new Agent id.
  </Card>
</CardGroup>

## Related

<CardGroup cols={2}>
  <Card title="Sessions" icon="clock" href="/agent-builder/sessions">
    Understand Builder session status, stopping a build, and the handoff to the saved Agent.
  </Card>

  <Card title="Run lifecycle" icon="circle-play" href="/concepts/runs">
    See what happens after the Agent starts a Run and produces outputs.
  </Card>
</CardGroup>