OpenClaw Browser Relay & Chrome Extension

What Is the OpenClaw Browser Relay?

The OpenClaw browser relay is a bridge between your AI agent and a real web browser. It lets your agent browse websites, read page content, click buttons, fill forms, take screenshots, and perform web-based tasks — all through natural language commands.

The browser relay works alongside the Chrome extension (also called ClawdBot browser relay) to give your agent real browser capabilities beyond simple web scraping.

How It Works

1. You install the Chrome extension in your browser. It connects to your OpenClaw instance via WebSocket.
2. When your AI agent needs to browse the web, it sends commands through the relay to the Chrome extension.
3. The extension executes the actions in a real Chrome tab — navigating pages, reading content, clicking elements, filling forms.
4. Results (page content, screenshots, action confirmations) are sent back to the agent.

Step 1: Enable the Browser Relay Skill

The browser relay is a ClawHub skill. On OpenClaw Launch, you can enable it from the configurator or your Dashboard page. For self-hosted setups, install the skill via the /skill install command.

Step 2: Install the Chrome Extension

1. Open Chrome and go to the Chrome Web Store.
2. Search for "ClawdBot browser relay" or "OpenClaw browser relay".
3. Click Add to Chrome and confirm the permissions.
4. Click the extension icon in the toolbar and enter your OpenClaw gateway URL (e.g., https://your-instance.openclawlaunch.com) and auth token.
5. Click Connect. The extension icon turns green when connected.

Step 3: Use Browser Commands

Once connected, you can ask your AI agent to perform browser tasks in natural language:

• "Go to example.com and summarize the page"
• "Search Google for the latest news about AI"
• "Fill out the contact form on this page"
• "Take a screenshot of this webpage"
• "Click the Sign Up button on the current page"

The agent interprets your request and translates it into browser actions via the relay.

What the Browser Extension Can Do

• Navigate to URLs and follow links
• Read and extract page content (text, tables, lists)
• Click buttons, links, and interactive elements
• Fill and submit forms
• Take screenshots of pages
• Scroll through pages and read long content
• Interact with JavaScript-rendered single-page applications

Permissions

The Chrome extension requests permissions to read and modify web pages. This is necessary for the agent to interact with page content. The extension only acts when your OpenClaw agent sends a command — it does not collect or send data on its own.

Browser Relay vs. Built-in Web Search

OpenClaw has a built-in web search skill that fetches and reads web pages without a browser extension. The browser relay adds interactive capabilities — clicking, form filling, JavaScript rendering, and real browser sessions. Use:

• Built-in web search — For reading articles, getting information, basic scraping
• Browser relay — For interactive tasks, SPAs, form filling, screenshots

Skip the Setup

On OpenClaw Launch, the browser relay skill is available as a one-click install. Deploy your AI agent first, then enable browser relay from your Dashboard page.