faster-chrome-devtools-skill

Controls Chrome via DevTools Protocol for snapshots, clicks, navigation, and screenshots.

4|Updated May 16, 2026
One-click install
npx skills add https://github.com/reason-machines/devtools-skills --skill faster-chrome-devtools-skill-reason-machines
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: faster-chrome-devtools-skill
Source: https://github.com/reason-machines/devtools-skills/tree/main/skills/faster-chrome-devtools-skill
Command: npx skills add https://github.com/reason-machines/devtools-skills --skill faster-chrome-devtools-skill-reason-machines

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve? Automating browser tasks usually requires heavy frameworks like Puppeteer or Playwright. This Skill controls Chrome directly through the Chrome DevTools Protocol (CDP) with a lightweight Node.js CLI, letting you navigate pages, interact with elements, capture screenshots, and inspect console or network errors without extra dependencies. ## Core Features & Use Cases - Accessibility Snapshots: Get a compact accessibility tree with stable element references for reliable clicking and form filling. - Navigation & Waits: Navigate to URLs with explicit timeouts, wait for text or selectors, and manage multiple tabs. - Screenshots & Debugging: Capture JPEG, PNG, or WebP screenshots (including full-page) and inspect console messages and failed network requests. - Flexible Connections: Attach to your logged-in Chrome, launch an anonymous Chromium instance, or connect to remote endpoints like Cloudflare Browser Run. - Use Case: Automate a login flow by navigating to a login page, filling credentials by selector, submitting the form, waiting for the dashboard text, and capturing a verification screenshot. ## Quick Start Ask the agent to navigate to a URL, take an accessibility snapshot, and click a specific element in Chrome.

Frequently Asked Questions about faster-chrome-devtools-skill

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

FAQPage Schema
How do I automate Chrome browser tasks without Puppeteer?▼

Use the Chrome DevTools Protocol directly via the cdp.mjs CLI script, which supports navigation, clicking, form filling, and screenshots. It requires only Node.js and connects to Chrome over a WebSocket with no additional dependencies.

How to take a screenshot of a webpage from the command line?▼

Run node scripts/cdp.mjs screenshot --path output.jpg after navigating to the page. The tool supports JPEG, PNG, and WebP formats, plus a --full-page flag to capture content below the fold.

Can I connect to my existing logged-in Chrome session?▼

Yes, launch Chrome with --remote-debugging-port=9222 and enable remote debugging at chrome://inspect/#remote-debugging. The CLI then attaches to your running instance, preserving cookies and login state.

Does this work with remote or cloud browser instances?▼

Yes, it supports remote endpoints via --ws-endpoint or --http-endpoint flags. For Cloudflare Browser Run, set CLOUDFLARE_ACCOUNT_ID and CLOUDFLARE_API_TOKEN environment variables and the CLI adds authentication headers automatically.

Why do I get connection refused when connecting to Chrome?▼

Connection refused means Chrome is not running with --remote-debugging-port=9222 or a firewall blocks localhost:9222. Verify the port appears at chrome://inspect/#remote-debugging and restart Chrome with the debugging flag.

What are the limitations of the background daemon connection?▼

The daemon times out after 20 minutes of inactivity, after which any command reconnects automatically. Stale daemons can be cleaned up with node scripts/cdp.mjs stop, optionally targeting a specific ID or endpoint.