ck:agent-browser

Automate browser sessions for AI agents using a snapshot and refs paradigm.

1|1|Updated Jan 15, 2026
One-click install
npx skills add https://github.com/Thanh-apero/apero-kit-cli --skill ck-agent-browser-thanh-apero
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ck:agent-browser
Source: https://github.com/Thanh-apero/apero-kit-cli/tree/main/.claude/skills/agent-browser
Command: npx skills add https://github.com/Thanh-apero/apero-kit-cli --skill ck-agent-browser-thanh-apero

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires playwright, puppeteer, chromium, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill addresses the challenge of long-running, autonomous browser automation for AI agents, providing a solution for tasks like video recording, self-verifying workflows, and cloud browser testing.

Core Features & Use Cases

  • Browser Automation: Automate browser actions such as navigation, snapshotting, and interaction with web elements.
  • Context Efficiency: Offers a 93% reduction in context compared to Playwright MCP, ideal for AI agents.
  • Use Case: Ideal for AI agents that require long-term browser sessions, such as automated testing, content analysis, and interactive web scraping.

Quick Start

Install the agent-browser skill globally and navigate to a URL.

Frequently Asked Questions about ck:agent-browser

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

FAQPage Schema
How do I reduce context usage for AI agent browser automation during long sessions?

You can reduce context usage for AI agent browser automation by using a snapshot and refs paradigm, which cuts context consumption by 93% compared to Playwright MCP, enabling efficient long-term autonomous sessions.

What is the best way to automate interactive web scraping and video recording with AI agents?

The best way to automate video recording and interactive web scraping with AI agents is to use a dedicated browser automation skill that manages Chromium sessions, allowing self-verifying workflows and content analysis.

Do I need to install Chromium separately to run cloud browser testing for AI agents?

Yes, you must install Chromium and optional system dependencies on Linux to run cloud browser testing, as the automation framework requires these underlying browser engines to execute sessions.

How does Playwright MCP compare to snapshot-based browser automation for autonomous tasks?

Playwright MCP consumes significantly more context than snapshot-based browser automation, which uses a snapshot and refs paradigm to achieve a 93% context reduction, making it superior for long-term autonomous tasks.

Why does my AI agent lose context awareness during long-running browser automation tasks?

AI agents lose context awareness during long-running browser automation because standard tools flood the context window with page data, whereas using a snapshot and refs paradigm reduces context by 93% to maintain session longevity.

Can I use Puppeteer and Playwright together for AI-driven browser automation?

Yes, this browser automation skill leverages both Puppeteer and Playwright as dependencies to drive Chromium, enabling AI agents to perform navigation, snapshotting, and web element interaction efficiently.