agent-browser

Automate browser interactions via CLI for testing, scraping, and screenshots.

3|2|Updated Mar 13, 2026
One-click install
npx skills add https://github.com/yoonhwan/ai-feature-pack --skill agent-browser-yoonhwan
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/yoonhwan/ai-feature-pack/tree/main/feature-pack/agent-browser/skill
Command: npx skills add https://github.com/yoonhwan/ai-feature-pack --skill agent-browser-yoonhwan

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Manually browsing websites, scraping data, taking screenshots, filling forms, and running end-to-end tests is repetitive, error-prone, and inefficient, especially when scaling across multiple pages or recurring workflows.

Core Features & Use Cases

  • Dual-mode browser automation: Supports fast native Rust CDP (no Node.js required) and stable Playwright mode for complex JavaScript single-page apps, with isolated sessions for parallel task execution.
  • Comprehensive interaction toolkit: Handles element clicks, form fills, network request mocking, cookie persistence, video recording, and console/network debugging for full end-to-end workflow automation.
  • Real-world use case: Automate login flows for internal tools, scrape product pricing data from e-commerce sites, or generate styled screenshot cards for team communications without manual browser interaction.

Quick Start

Instruct your AI agent to navigate to a target webpage, fill in a provided login form with test credentials, and capture a screenshot of the resulting dashboard to verify the login flow works correctly.

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 scraping and form filling without writing complex browser automation scripts?

You can automate web scraping and form filling by instructing an AI agent to use the agent-browser CLI, which handles page navigation, element clicking, and form interactions via a command-line interface without requiring complex scripting.

Does this browser automation tool support complex JavaScript single-page applications?

Yes, browser automation supports complex JavaScript single-page applications through Playwright mode, while also offering a native Rust CDP mode for faster performance on standard pages without requiring Node.js.

What do I need to start running end-to-end testing and login flow automation from the command line?

You need the agent-browser CLI version 0.16.3 or higher installed globally on your system to start running end-to-end testing, login flow automation, and other web interaction tasks from the command line.

Can I capture screenshots and record video during web browser automation sessions?

Yes, you can capture screenshots and record video during web browser automation sessions, and you can also mock network requests, persist cookies, and debug console or network issues within isolated sessions.

What is the best way to generate styled visual cards from HTML content for team communications?

The best way to generate styled visual cards from HTML content is to use a CLI browser automation tool that captures styled screenshots of web pages, eliminating the need for manual browser interaction when creating visual assets.

When should I use native Rust CDP mode over Playwright mode for web scraping?

You should use the native Rust CDP mode for improved performance and faster execution when Node.js is unavailable, and switch to Playwright mode when dealing with complex JavaScript single-page applications that require advanced rendering.