import { CodeBlock } from '../views/docs/CodeBlock';
import { NextLink, Callout } from '../views/docs/prose';

## Quickstart

<Callout>
The recommended way to build Guava voice agents is using the Guava CLI, which bootstraps projects and installs the SDK.
If you’d rather not install the CLI, you can skip to the [direct SDK installation guide](./sdk-installation) instead.
</Callout>

### Create an account

Sign up at [app.goguava.ai](https://app.goguava.ai).

### Install the CLI

Pick an install method below.

<CodeBlock code={`# (macOS / Linux) Use the install script. Installs to \`~/.local/bin/guava\` 
curl -fsSL https://storage.googleapis.com/gridspace-guava-cli/cli/install.sh | sh

# (macOS) Install using Homebrew
brew tap goguava-ai/tap
brew install goguava-ai/tap/guava`} language="bash" />

### Authenticate the CLI

<CodeBlock code="guava login" language="bash" />

### Create an Agent

Scaffold a new agent project with starter code (currently Python only).

<CodeBlock code="guava create my-agent" language="bash" />

### Add vibe-coding kit (optional)

Clone the Guava starter repository into your new agent project. It contains plain-text API docs and examples sized for AI coding assistants.

<CodeBlock code={`git clone https://github.com/goguava-ai/guava-starter.git my-agent/guava-starter`} language="bash" />

### Deploy your Agent

<CodeBlock code="guava deploy up ./my-agent" language="bash" />

Track status in [Deployments](https://app.goguava.ai/dashboard/deployments). Every call appears in [Conversations](https://app.goguava.ai/dashboard/conversations).

<NextLink section="inbound-rag-example" label="Inbound Example w/ RAG" />