chrome-cdp

Control a local Chrome session via the Chrome DevTools Protocol from the command line.

6|1|Updated Nov 23, 2023
One-click install
npx skills add https://github.com/gmcabrita/dotfiles --skill chrome-cdp-gmcabrita
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: chrome-cdp
Source: https://github.com/gmcabrita/dotfiles/tree/main/.agents/skills/chrome-cdp
Command: npx skills add https://github.com/gmcabrita/dotfiles --skill chrome-cdp-gmcabrita

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Developers need a lightweight, Puppeteer-free way to script and automate interactions with a local Chrome session using the Chrome DevTools Protocol.

Core Features & Use Cases

  • List open pages and targets, inspect page metadata, and manage per-tab sessions
  • Take viewport screenshots, evaluate JavaScript, navigate, and interact with elements across multiple tabs
  • Use the per-tab daemon to keep a persistent CDP connection and script complex workflows safely

Quick Start

Start by listing targets with cdp list, then capture a screenshot with cdp shot <target>.

Frequently Asked Questions about chrome-cdp

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

FAQPage Schema
How do I control a local Chrome session via the Chrome DevTools Protocol from the command line?

You can control a local Chrome session via the Chrome DevTools Protocol from the command line by using a lightweight per-tab daemon and CLI scripts to navigate pages, evaluate JavaScript, and take screenshots without needing Puppeteer.

Do I need Puppeteer to script Chrome automation and interact with elements across multiple tabs?

No, you do not need Puppeteer to script Chrome automation. This approach uses a lightweight per-tab daemon to maintain a persistent CDP connection, allowing you to interact with elements across multiple tabs directly from the CLI.

What is required to automate web interactions using the Chrome DevTools Protocol?

Automating web interactions using the Chrome DevTools Protocol requires a local Chrome session with remote debugging enabled and Node.js 22 or higher to run the CLI scripts and manage the per-tab daemon.

How do I list open pages and take viewport screenshots using CDP?

To list open pages and take viewport screenshots using CDP, you can run CLI commands to list available targets, then specify a target to capture a screenshot of its viewport.

Can I evaluate JavaScript and navigate pages across multiple tabs without a browser automation library?

Yes, you can evaluate JavaScript and navigate pages across multiple tabs without a browser automation library by sending commands directly through the Chrome DevTools Protocol using a lightweight CLI script.