agent-browser

Automate browser navigation, element interaction, and screenshot capture in Chromium.

Updated Feb 28, 2026
One-click install
npx skills add https://github.com/raphaelmans/agent-skills --skill agent-browser-raphaelmans
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/raphaelmans/agent-skills/tree/main/agent-browser
Command: npx skills add https://github.com/raphaelmans/agent-skills --skill agent-browser-raphaelmans

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Automates repetitive browser-based tasks such as navigating pages, interacting with elements, taking screenshots, and extracting information, reducing manual effort and human error.

Core Features & Use Cases

  • Automates end-to-end web workflows including navigation, element interaction, data capture, and visual verification.
  • Supports refs-based interactions and semantic locators for robust automation across dynamic pages.
  • Ideal for QA testing, form automation, data extraction, and documentation/demo recording.

Quick Start

Open a URL with agent-browser, snapshot the page, and interact with elements using refs to fill forms and capture results.

Frequently Asked Questions about agent-browser

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

FAQPage Schema
How do I automate browser interactions for web testing and form filling?

You can automate browser interactions by navigating to a URL, snapshotting the page, and interacting with elements using semantic locators and refs to execute reliable, repeatable web workflows like form filling and testing.

Can I extract data and capture screenshots during web page navigation?

Yes, data extraction and screenshot capture are supported natively during web page navigation. The tool snapshots page state and captures visual recordings, enabling automated data extraction and visual verification for QA workflows.

Do I need a specific browser environment to run web automation workflows?

Yes, you need a Chromium-based environment to run these web automation workflows. This environment requirement ensures reliable execution of navigation, element interaction, and state capture commands across dynamic pages.

What is the best way to interact with dynamic web pages during automated testing?

The best way to interact with dynamic web pages is using refs-based interactions and semantic locators. This approach provides robust automation by targeting elements reliably, even as page content changes during navigation and testing.

How does state capture and page snapshotting work for automated data extraction?

State capture and page snapshotting work by recording the current page state and element references during navigation. This exposes commands for snapshotting, interaction, and waiting, which streamlines automated data extraction and visual verification.

Can I use browser automation for end-to-end testing and demo documentation?

Yes, browser automation is ideal for end-to-end testing and demo documentation. It supports navigation, element interaction, data capture, and visual recording, allowing you to automate repetitive workflows and generate documentation or demonstrations.