agent-browser

Automate web browser interactions via CLI with snapshots and element references.

Updated Jan 27, 2026
One-click install
npx skills add https://github.com/Hoang130203/Emgu_CV_automation --skill agent-browser-hoang130203
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/Hoang130203/Emgu_CV_automation/tree/main/.opencode/skills/agent-browser
Command: npx skills add https://github.com/Hoang130203/Emgu_CV_automation --skill agent-browser-hoang130203

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill streamlines browser automation for AI agents, significantly reducing context window usage and enabling efficient, long-running autonomous sessions.

Core Features & Use Cases

  • Context-Efficient Snapshots: Uses a "snapshot + refs" paradigm for drastically reduced context compared to traditional tools like Playwright.
  • Autonomous Sessions: Ideal for long, self-verifying workflows where AI needs to interact with web pages over extended periods.
  • Cloud Browser Testing: Seamless integration with cloud providers like Browserbase for CI/CD and remote testing.
  • Video Recording: Built-in functionality to record browser sessions for debugging.

Quick Start

Navigate to example.com, snapshot the page for interactive elements, and then click the element referenced as @e1.

Frequently Asked Questions about agent-browser

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

FAQPage Schema
How do I reduce context window usage during AI agent browser automation?

To reduce context window usage for AI agent browser automation, use a snapshot and reference system. This approach captures interactive elements as references, drastically cutting token consumption compared to traditional Playwright scripts.

Can I run autonomous web sessions with Playwright and an AI agent CLI?

Yes, you can run autonomous web sessions using this AI-optimized CLI. It supports navigation, element analysis, and state management, enabling long-running, self-verifying workflows for AI agents interacting with web pages.

Does this browser automation CLI support cloud testing with Browserbase?

Yes, this browser automation CLI supports seamless integration with cloud providers like Browserbase. This enables remote browser testing and CI/CD workflows for autonomous web sessions.

What is the best way to interact with web elements using an AI agent?

The best way to interact with web elements using an AI agent is capturing a page snapshot and referencing elements directly. You can click elements referenced as @e1, providing fine-grained control via extensive CLI command options.

How do I record browser sessions for debugging web scraping workflows?

You can record browser sessions for debugging web scraping workflows using the CLI's built-in video recording functionality. This captures media during autonomous web sessions for later review and troubleshooting.

What are the limitations of using a snapshot system for web scraping?

The snapshot system for web scraping relies on captured interactive elements and references, which may require frequent updates if the page DOM changes dynamically during autonomous sessions, potentially interrupting element interaction.