js-eyes

Manage browser tabs, extract HTML, and run scripts via WebSocket.

43|7|Updated Jan 19, 2026
One-click install
npx skills add https://github.com/imjszhang/js-eyes --skill js-eyes
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: js-eyes
Source: https://github.com/imjszhang/js-eyes/tree/main
Command: npx skills add https://github.com/imjszhang/js-eyes --skill js-eyes

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

JS Eyes connects AI agent frameworks to real web browsers through a WebSocket bridge so they can manage tabs, extract page content, execute scripts, and handle cookies without fragile manual automation.

Core Features & Use Cases

  • Persistent WebSocket Bridge: Extension clients maintain live connections to a Node.js server that exposes tab and browser metadata for agent use.
  • OpenClaw Toolset & CLI: The OpenClaw plugin registers multiple JS Eyes tools plus service and CLI commands, enabling agents to query tabs, open URLs, extract HTML, run scripts, and list clients or cookies.
  • Extension Skills Ecosystem: Additional skills like js-search-x can be discovered and installed to extend capabilities such as X.com scraping, creating a modular automation stack.

Quick Start

Ask your agent to connect to the JS Eyes server, list all browser tabs, and fetch the HTML for a target page.

Frequently Asked Questions about js-eyes

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I give AI agents real browser automation capabilities for tab management?

You can give AI agents browser automation by connecting them to a local WebSocket bridge that exposes HTTP endpoints for tab orchestration, content extraction, and script execution. This allows frameworks to deterministically control live browser sessions.

Can I use WebSocket browser automation to extract HTML and run scripts with OpenClaw?

Yes, WebSocket browser automation supports OpenClaw by registering tools and CLI commands that allow agents to open URLs, extract page HTML, execute scripts, and list connected clients through a persistent Node.js server.

Does AI agent browser automation require a Node.js server and browser extensions?

AI agent browser automation requires a bundled Node.js server, browser extensions, and the ws dependency to maintain live WebSocket connections and expose consistent HTTP status endpoints for agent control.

What's the best way to manage browser cookies with AI agents through a local service?

The best way to manage cookies with AI agents is through a local WebSocket service that registers tool commands, enabling frameworks to access and list browser cookies alongside managing tabs and executing scripts.

How do I monitor connected browser clients for AI agent automation?

You monitor connected browser clients by querying the local Node.js server through exposed CLI commands, which lists all active extension clients maintaining live WebSocket connections for agent orchestration.

Can I extend browser automation skills to scrape specific platforms like X.com?

Yes, you can extend browser automation by discovering and installing additional skills like js-search-x, which adds platform-specific scraping capabilities to create a modular automation stack for AI agents.