Get started / Getting started

Build your first voice agent

From zero to a live agent answering calls in four steps — grab a key, pick a build path, test in the Playground, and attach a number.

Quickstart

Four steps to your first call.

1. 1

   Grab an API key

   Dashboard → Settings → API keys. Keys are environment-scoped — dev, test, prd, live.

2. 2

   Pick your build path

   Flow builder to compose on canvas, the SDK to write an agent in code, or a brain connector to put a voice on an existing LLM app. All three land in the same Agents list.

3. 3

   Test in the Playground

   Talk to your agent from the browser — voice or text. No phone number needed.

4. 4

   Attach a number

   Buy a number under Numbers, point it at your agent, and dial in — it answers and runs the process.

Write the agent SDK

On the code path, install @voicelayer/sdk — then this is a complete agent. Prefer clicks? The flow builder compiles to the same runtime.

import { Agent } from '@voicelayer/sdk';

export default await new Agent({
  name: 'fnol-simple',
  prompt: 'You are a calm First Notice of Loss intake agent.',
  capabilities: ['fnol', 'inbound-sip'],
  tools: {
    transferToHuman: {
      description: 'Transfer the caller to a human claims agent.',
      input: { reason: 'string' },
      run: async ({ reason }, ctx) => {
        await ctx.handoff('+15555550100', { reason });
        return 'transferred';
      }
    }
  }
}).start(import.meta.url);

Your default models

Every call runs a STT → LLM → TTS pipeline. These are the defaults — override per-project (BYOK) or per-agent.