browser

Control Chrome via CDP to navigate, evaluate scripts, and capture screenshots.

1|Updated Mar 8, 2026
One-click install
npx skills add https://github.com/tense-i/stock-market-simulator --skill browser-tense-i
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: browser
Source: https://github.com/tense-i/stock-market-simulator/tree/main/skills/browser
Command: npx skills add https://github.com/tense-i/stock-market-simulator --skill browser-tense-i

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires ws, and includes scripts (resource) components.

What problem does it solve?

Automates browser automation tasks by leveraging the Chrome DevTools Protocol (CDP) to control a running Chrome instance locally, enabling scripted navigation, JS evaluation, element inspection, and screenshots without a full automation framework.

Core Features & Use Cases

  • CDP-powered navigation: Launch Chrome, navigate to URLs, switch tabs, and execute JavaScript in the page context.
  • Automation tasks on demand: Capture screenshots, inspect DOM nodes, and collect metadata for testing or QA.
  • Use Case: Quickly automate a QA scenario where you navigate to a page, verify the title, and save a screenshot for reports.

Quick Start

Launch Chrome with remote debugging on port 9222 and run the helper scripts to perform a sequence of CDP-based actions.

Frequently Asked Questions about browser

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

FAQPage Schema
How do I automate browser tasks with Chrome DevTools Protocol locally?

Automate browser tasks with CDP by connecting to a running Chrome instance over localhost:9222 to navigate pages, evaluate scripts, take screenshots, and inspect DOM elements without a dedicated server.

What's the best way to execute JavaScript in a page context using CDP?

Execute JavaScript in a page context using CDP by connecting to a running Chrome instance locally and running helper scripts that evaluate commands directly in the browser environment and return structured JSON results.

Can I take screenshots and inspect DOM nodes without a full automation framework?

Take screenshots and inspect DOM nodes without a full automation framework by using CDP helper scripts to control a local Chrome instance, capturing page images and collecting metadata for testing or QA reports.

Do I need a dedicated MCP server to connect to Chrome over localhost:9222?

No dedicated MCP server is needed to connect to Chrome over localhost:9222. You can launch Chrome with remote debugging enabled and use CDP helper scripts directly to perform interactive browser automation.

How does CDP-based navigation handle tab switching and URL routing in automation?

CDP-based navigation handles tab switching and URL routing by sending protocol commands over a WebSocket connection to a local Chrome instance, enabling scripts to launch Chrome, navigate to URLs, and switch tabs programmatically.

What are the limitations of using CDP over WebSocket for local browser automation?

Limitations of using CDP over WebSocket include requiring a manually launched Chrome instance with remote debugging on port 9222, relying on the ws dependency, and lacking the built-in driver management found in full automation frameworks.