chrome-cdp-skill

Control live Chrome tabs via the Chrome DevTools Protocol.

Updated Mar 18, 2026
One-click install
npx skills add https://github.com/okbest2024/openclaw-backup --skill chrome-cdp-skill
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: chrome-cdp-skill
Source: https://github.com/okbest2024/openclaw-backup/tree/main/workspace/skills/chrome-cdp
Command: npx skills add https://github.com/okbest2024/openclaw-backup --skill chrome-cdp-skill

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Chrome sessions that are already open can be read and controlled by AI agents through the Chrome DevTools Protocol, enabling real-time inspection and interaction without relaunching a browser.

Core Features & Use Cases

  • List and view all open Chrome tabs and their URLs using a lightweight per-tab daemon.
  • Retrieve HTML, accessibility trees, and inspect page state; take screenshots for visual verification.
  • Interact with pages by clicking elements, typing text, navigating to URLs, and evaluating JavaScript within the active tab.
  • Compare against OpenClaw Browser tooling and debug live sessions for tasks like QA, data extraction, or automation scenarios.

Quick Start

Clone the chrome-cdp-skill repository, enable remote debugging in Chrome, and place the chrome-cdp directory into your agent’s skills folder.

Frequently Asked Questions about chrome-cdp-skill

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

FAQPage Schema
How do I automate live Chrome tabs without relaunching the browser?

To inspect a live Chrome session, you must enable remote debugging in Chrome and have a Node.js runtime available to execute the skill's internal scripts that communicate over WebSocket CDP endpoints.

Can I capture screenshots and retrieve HTML from open Chrome tabs via CDP?

Yes, connecting to Chrome via CDP allows you to retrieve HTML, capture accessibility trees, inspect page state, and take screenshots for visual verification of the active tab.

How do I click elements and evaluate JavaScript in a Chrome tab using an AI agent?

By communicating over the Chrome DevTools Protocol WebSocket, an AI agent can interact with pages by clicking elements, typing text, navigating to URLs, and evaluating JavaScript within the target tab.

What is the best way to debug live browser sessions for QA automation?

Using the Chrome DevTools Protocol to connect to running Chrome sessions is an effective approach for QA automation, enabling real-time inspection, interaction, and data extraction without needing to relaunch the browser.