Quick Start Guide

This guide will walk you through installing and configuring Nanobrowser for the first time.

Step 1: Install Nanobrowser

If you haven't already, install the Nanobrowser extension from the Chrome Web Store.

Install from Chrome Web Store

For more installation options, see the Installation page.

Step 2: Open the Side Panel

Once installed, click the puzzle piece icon in your Chrome toolbar to open your extensions list, and then click on Nanobrowser. This will open the Nanobrowser side panel.

For easier access, you can pin the extension by clicking the pin icon next to its name.

Step 3: Configure Your LLM Provider

Nanobrowser requires you to use your own API keys for Large Language Model (LLM) providers. Let's configure one.

  1. In the Nanobrowser side panel, click the Settings icon (⚙️) in the top-right corner.
  2. In the LLM Providers section, click the + Add New Provider button.
  3. Select a provider from the list, for example, OpenAI.
  4. Enter your API Key in the input field.
  5. Click Save.

Your provider is now configured and ready to use.

Step 4: Assign Models to Agents

Nanobrowser uses a multi-agent system to perform tasks. You need to assign a model to each agent.

  • Planner: The strategic agent that breaks down your task into high-level steps.

  • Navigator: The execution agent that interacts with the webpage (clicking, typing, etc.).

  • In the settings page, scroll down to the Model Selection section.

  • For the Planner agent, select your newly configured provider and a powerful model (e.g., gpt-4o).
  • For the Navigator agent, you can use the same model or a faster, more cost-effective one (e.g., gpt-4o-mini if available, or the same gpt-4o).

For more detailed recommendations, see the Choosing Your Models guide.

Step 5: Run Your First Task

Now you're ready to automate! Go back to the main chat interface.

  1. Navigate to a website, for example, https://techcrunch.com/.

  2. In the Nanobrowser chat input at the bottom of the side panel, type your command:

    Go to TechCrunch and extract the top 10 headlines from the last 24 hours.

  3. Press Enter or click the Send button.

Watch as Nanobrowser's agents work together to understand the page and execute your command. You'll see real-time updates in the chat panel as the agents report their actions and findings.

Congratulations, you've successfully run your first AI-powered web automation task!