agent-browser

Automate browser interactions and scrape web pages via the agent-browser CLI.

8|4|Updated Jan 15, 2026
One-click install
npx skills add https://github.com/jscraik/Agent-Skills --skill agent-browser-jscraik
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/jscraik/Agent-Skills/tree/main/utilities/agent-browser
Command: npx skills add https://github.com/jscraik/Agent-Skills --skill agent-browser-jscraik

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill provides deterministic browser automation and scraping capabilities through a command-line interface, enabling reliable interaction with web pages and extraction of their state.

Core Features & Use Cases

  • Deterministic Browser Automation: Interact with live web pages, fill forms, and click elements using accessibility-based references for stability.
  • Snapshotting and Scraping: Capture screenshots, PDFs, and structured snapshots of page states for verification and data extraction.
  • Use Case: Automate the process of filling out a complex online form by providing a URL and a sequence of actions, ensuring each step is verified against a fresh snapshot of the page.

Quick Start

Use the agent-browser skill to open 'https://example.com', take a snapshot, and click the element with ref '@submit-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 form submissions and capture UI states via command line?

Deterministic browser automation executes web interactions via the agent-browser CLI, using accessibility-based ref snapshots to click elements and fill forms with verifiable, stable steps.

What is ref-based snapshotting for web scraping and how does it work?

Ref-based snapshotting for web scraping captures structured page states using accessibility references, allowing the browser automation CLI to target and interact with elements stably across navigation steps.

Do I need to install the agent-browser CLI tool before using this browser automation skill?

Yes, using this CLI browser automation skill requires the agent-browser CLI tool to be installed and functional locally to execute navigation, snapshotting, and interaction tasks.

Can I extract data from web pages and verify each automation step against a fresh snapshot?

Yes, you can extract data from web pages and verify each step by taking a fresh snapshot of the page state before executing actions like clicking or filling, ensuring deterministic scraping.

Does CLI browser automation work for capturing screenshots and PDFs of live web pages?

Yes, CLI browser automation works for capturing screenshots and PDFs of live web pages, enabling you to verify UI states and extract data through deterministic, scriptable actions.