Chrome Cdp

A lean CDP client that lets Claude inspect and control your actual Chrome tabs without spinning up Puppeteer or a headless browser. You run commands to list open tabs, grab screenshots, dump accessibility trees, or eval JavaScript against whatever's already running locally. Works across 100+ tabs, instant WebSocket connection. The coordinate system is worth noting: screenshots are at native resolution (CSS pixels times DPR) but click commands take CSS pixels, so you'll divide by DPR when translating positions. Useful when you want Claude to debug a live page, grab state from an existing session, or automate something in a tab you already have open. Chrome shows a one-time allow prompt per tab, then a daemon keeps the session alive.

skills.sh installs: 2.1K
GitHub stars: 3.3K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add pasky/chrome-cdp-skill --skill chrome-cdp --agent claude-code

Installs into .claude/skills of the current project.

Files

Chrome CDP

Lightweight Chrome DevTools Protocol CLI. Connects directly via WebSocket — no Puppeteer, works with 100+ tabs, instant connection.

Prerequisites

Chrome (or Chromium, Brave, Edge, Vivaldi) with remote debugging enabled: open chrome://inspect/#remote-debugging and toggle the switch
Node.js 22+ (uses built-in WebSocket)
If your browser's DevToolsActivePort is in a non-standard location, set CDP_PORT_FILE to its full path

Commands

All commands use scripts/cdp.mjs. The <target> is a unique targetId prefix from list; copy the full prefix shown in the list output (for example 6BE827FA). The CLI rejects ambiguous prefixes.

List open pages

scripts/cdp.mjs list

Take a screenshot

scripts/cdp.mjs shot <target> [file]    # default: screenshot-<target>.png in runtime dir

Captures the viewport only. Scroll first with eval if you need content below the fold. Output includes the page's DPR and coordinate conversion hint (see Coordinates below).