Local quickstart

Build a working agent app on your machine with a direct model provider. This tutorial does not require a Veryfront account or Veryfront Cloud.

Prerequisites

  • Node.js 22.3 or later.
  • An OpenAI API key for model inference.

Create the app

bash
npm create veryfront@latest support-agent -- --template ai-agent
cd support-agent

The starter contains an assistant agent, calculator.ts tool, chat page, AG-UI route, and smoke eval.

Answer No to Use Veryfront Cloud? to follow this direct-provider tutorial. To use Veryfront inference and inspect local runs in Studio without a provider API key, follow the Cloud quickstart.

Configure inference

Set the API key in the terminal where you run Veryfront:

bash
export OPENAI_API_KEY="<API_KEY>"

The starter uses openai/gpt-5-nano. Veryfront sends the requests directly to OpenAI.

Run the app

bash
npm run dev

The CLI confirms the server and available inference path:

text
✓ App ready       http://localhost:3000
  Execution       This machine
Inference OpenAI direct

Verify it worked

Open the local URL printed by the CLI and ask:

text
What is 128 divided by 8?

The assistant calls the calculator tool and returns 16.

Run the eval

Stop the dev server with Ctrl+C, then run:

bash
npm run eval -- assistant

The command exits successfully after the agent calls the calculator and returns the expected answer.

Add local delegation

Continue with Local delegation when you need more than one agent.

Next steps

Use another inference provider when you need a different model or local inference.