ClawKit Logo
ClawKitReliability Toolkit
Back to Registry
Official Verified browser automation Safety 2/5

js-eyes

Browser automation for AI agents — control tabs, extract content, execute scripts and manage cookies via WebSocket.

Why use this skill?

Control your browser with AI. JS Eyes enables OpenClaw agents to manage tabs, extract content, execute scripts, and handle cookies seamlessly via WebSocket.

skill-install — Terminal

Install via CLI (Recommended)

clawhub install openclaw/skills/skills/imjszhang/js-eyes
Or

What This Skill Does

JS Eyes acts as a bridge between your AI agent and the browser environment. It enables comprehensive browser automation by maintaining a WebSocket connection between the browser extension and an OpenClaw server. Through this connection, the agent gains direct control over browser activities including navigation, content extraction, and scripting. It effectively treats the browser as an API, allowing your AI to perform complex workflows such as web scraping, automated form submission, and real-time page analysis without manual intervention. By managing tabs, cookies, and JavaScript execution, it provides the full feature set required for deep web interaction.

Installation

To integrate JS Eyes, ensure you have Node.js 16 or higher installed on your local machine. Install the skill directly via the ClawHub CLI using the following command:

clawhub install openclaw/skills/skills/imjszhang/js-eyes

Once installed, you must start the background server process using openclaw js-eyes server start. Ensure the corresponding browser extension is installed in your preferred browser (Chrome, Edge, or Firefox) to establish the initial connection to the WebSocket server.

Use Cases

JS Eyes is ideal for autonomous research, regression testing, and data aggregation. You can use it to build an agent that scans multiple news sites for specific stock movements, interacts with web applications to generate status reports, or automates tedious login and data retrieval flows that lack formal APIs. It is perfect for developers who need to bridge the gap between their local AI agent and dynamic web content.

Example Prompts

  1. "Open the URL https://www.github.com, extract all the trending repositories, and summarize the top three in a list."
  2. "Check my currently open tabs, find the one containing the project dashboard, and run a script to get the current system status from the DOM."
  3. "List the cookies for the current domain and tell me if the session cookie is set to expire within the next 24 hours."

Tips & Limitations

  • Security: Always verify the origins of the scripts you execute. Since js_eyes_execute_script runs arbitrary code in your browser context, ensure you are interacting with trusted pages.
  • Performance: Heavy DOM extraction can be resource-intensive for very large pages. If you are retrieving HTML, consider using execute_script to narrow down the scope to specific elements before requesting the full content.
  • Connection: The plugin depends on a persistent WebSocket connection. If the server process stops, the agent will lose its ability to interact with the browser. Always check openclaw js-eyes status if the agent reports connectivity errors.

Metadata

Author@imjszhang
Stars2287
Views1
Updated2026-03-09
View Author Profile
AI Skill Finder

Not sure this is the right skill?

Describe what you want to build — we'll match you to the best skill from 16,000+ options.

Find the right skill
Add to Configuration

Paste this into your clawhub.json to enable this plugin.

{
  "plugins": {
    "official-imjszhang-js-eyes": {
      "enabled": true,
      "auto_update": true
    }
  }
}

Tags(AI)

#browser-automation#web-scraping#automation#javascript#websocket
Safety Score: 2/5

Flags: network-access, code-execution, data-collection