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

# Tutorial: Create an AI CLI App with NodeJS and Tyk AI Studio

> Learn how to use your Tyk AI Studio App credentials to build a simple command-line interface (CLI) application with NodeJS.

In this tutorial, you will learn how to use the credentials from an App you've created in Tyk AI Studio to build a simple command-line interface (CLI) application with NodeJS.

This application will take a text prompt from you, send it to an LLM via the Tyk AI Gateway, and print the response.

## Prerequisites

Before you begin, you will need:

* **An approved App in Tyk AI Studio:** You should have already created an App and had it approved by a Studio Administrator.

  If you haven't, please follow the guide on [App Management for Consumers](/5.12/ai-management/ai-studio/ai-portal-app).
* **Your App's Secret:** You will need these to authenticate your application.
* **Node.js and npm installed:** This tutorial uses NodeJS. You can download it from [nodejs.org](https://nodejs.org/).
* **An LLM configured in AI Studio:** Your App must be configured to use at least one LLM. This tutorial will use OpenAI LLM models.

## Instructions

### Step 1: Set up your NodeJS project

First, create a new folder for your project and initialize a new NodeJS project within it.

```bash theme={null}
mkdir tyk-ai-cli-app
cd tyk-ai-cli-app
npm init -y
npm install ai @ai-sdk/openai
```

### Step 2: Create the CLI application file

Create a new file named `index.js` in your project folder. This is where you will write the code for your CLI application.

```bash theme={null}
touch index.js
```

### Step 3: Write the application code

Open `index.js` in your favorite code editor and add the following code. Be sure to replace the placeholder values with your actual App credentials and AI Gateway endpoint.

```javascript Expandable theme={null}
import readline from "node:readline";
import { generateText } from "ai";
import { createOpenAI } from "@ai-sdk/openai";

// ---- CONFIG ----
const apiKey = "YOUR_API_KEY";
const baseURL = "YOUR_TYK_AI_STUDIO_GATEWAY_URL"; 

// Create OpenAI-compatible client
const openai = createOpenAI({
  apiKey,
  baseURL
});

// CLI interface
const rl = readline.createInterface({
  input: process.stdin,
  output: process.stdout,
  prompt: "You> "
});

console.log("LLM CLI started. Type a prompt and press enter.\n");
rl.prompt();

rl.on("line", async (line) => {
  const prompt = line.trim();

  if (!prompt) {
    rl.prompt();
    return;
  }

  try {
    const { text } = await generateText({
      model: openai("gpt-4o-mini"),
      prompt
    });

    console.log("\nLLM>", text, "\n");
  } catch (err) {
    console.error("Error:", err.message);
  }

  rl.prompt();
});
```

### Step 4: Run your CLI App

Now you can run your application from your terminal.

```bash theme={null}
node index.js
```

Example:

```
LLM CLI started. Type a prompt and press enter.

You> explain kubernetes in simple terms

LLM> Kubernetes is a system that helps run and manage containers...
```

Congratulations! You have successfully used your Tyk AI Studio App credentials to build a working AI-powered CLI application.

## Frequently Asked Questions

<AccordionGroup>
  <Accordion title="I'm getting a 401 Unauthorized error. What's wrong?">
    A 401 error means your request is not properly authenticated. Check the following:

    * Make sure your App has been approved by an administrator and its credentials are active.
    * Double-check that you have correctly copied your Secret into the `index.js` file.
  </Accordion>

  <Accordion title="Can I use a different LLM?">
    Yes, you can use any LLM that you have configured in your App.
  </Accordion>
</AccordionGroup>