agent-browser

Automate Chrome/Chromium browser tasks via CDP for AI agents.

Updated Mar 23, 2026
One-click install
npx skills add https://github.com/fray-cloud/coin --skill agent-browser-fray-cloud
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/fray-cloud/coin/tree/main/.agents/skills/agent-browser
Command: npx skills add https://github.com/fray-cloud/coin --skill agent-browser-fray-cloud

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

AI agents often require interacting with websites to perform tasks such as login, data extraction, form submission, or testing. This skill provides a robust, scriptable browser automation CLI to replace manual browsing with deterministic, repeatable actions.

Core Features & Use Cases

  • Navigate, snapshot, and interact with web pages using a concise command set (open, snapshot, fill, click, wait, screenshot, etc.).
  • Use compact element refs from interactive snapshots to drive automation with minimal context, improving reliability and reducing token usage.
  • Supports advanced patterns like authentication flows, session management, device/emulation, proxies, and iframe handling for end-to-end web automation and testing.

Quick Start

Open a URL, snapshot the page to obtain element refs, and perform actions like filling a form or clicking a button.

Frequently Asked Questions about agent-browser

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

FAQPage Schema
How do I automate web tasks for AI agents using browser control?

You can automate web tasks for AI agents by using a scriptable CLI that controls Chrome or Chromium via the Chrome DevTools Protocol (CDP) to navigate, snapshot, and interact with web pages deterministically. This approach replaces manual browsing with repeatable actions for data extraction and testing.

How does browser automation handle authentication and session management?

Browser automation handles authentication and session management by supporting persistent sessions, proxies, and device emulation, allowing AI agents to log in, maintain authenticated states across sites, and execute chained commands without repeated manual interventions.

What is the best way to scrape data and fill forms via CDP?

The best way to scrape data and fill forms via CDP is to open a URL, generate an interactive snapshot to obtain compact element references, and use those references to drive actions like filling inputs or clicking buttons with minimal context and reduced token usage.

Can I use this browser automation CLI for testing web apps across different devices?

Yes, you can use this browser automation CLI for testing web apps across different devices because it supports device emulation, iframe handling, and proxy configurations to simulate various environments and ensure reliable end-to-end web testing.

Why use interactive snapshots for web scraping and element interaction?

Using interactive snapshots for web scraping provides compact element references that drive automation with minimal context. This improves reliability and reduces token usage when AI agents need to fill forms, click buttons, or extract data from web pages.