Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.yutori.com/llms.txt

Use this file to discover all available pages before exploring further.

The Navigator Browser Extension is a lightweight Chrome extension that wraps around the Navigator API to automate browser tasks. It’s a convenient way to try out both Navigator n1.5 and Navigator n1 on real pages without writing any code — Navigator n1.5 is the default and recommended model. Simply describe what you want to do, and the agent will execute it step-by-step—clicking, typing, scrolling, and navigating autonomously.

​
Prerequisites

​
Installation

  1. Download the Navigator Browser Extension
  2. Unzip the file to a folder on your computer
  3. Open Chrome and navigate to chrome://extensions/
  4. Enable Developer mode (toggle in the top right corner)
  5. Click Load unpacked (button on the top left corner)
  6. Select the unzipped extension folder
  7. The extension icon should appear in your toolbar
    • If it doesn’t appear, click the puzzle piece icon to see your list of extensions and pin it to your toolbar
Chrome extensions page showing Developer mode toggle (1), Load unpacked button (2), and puzzle piece icon (3)

​
Configure Your API Key

  1. Click the extension icon in your Chrome toolbar
  2. The settings page will automatically open the first time launching the extension
    • To change settings afterwards, click the gear icon to open settings
  3. Enter your Yutori API key
  4. Pick a model — Navigator n1.5 is selected by default and is the recommended choice for new users; switch to Navigator n1 to test the older model
  5. Click Save

​
How to Use

  1. Navigate to any webpage
  2. Click the extension icon
  3. Type your task in natural language, for example:

On AWS Console webpage

Create a new IAM role called “api-test” with API Gateway use-case and set the env tag to “dev”

On Hugging Face webpage

Create a new access token called “test-token” and give it only read access
  1. Press Enter to submit task
  2. Watch the agent work—it will show its thoughts and actions in real-time
    • You can also switch to a different tab and let the task run in the background
    • Multiple tasks can be launched in parallel across multiple tabs
  3. Click Stop at any time to pause execution
  4. Stopped or completed tasks will provide a summary of its progress and what was accomplished