agent-browser

Control a headless or headed browser via CLI for navigation, form filling, and screenshots.

3|Updated Dec 22, 2025
One-click install
npx skills add https://github.com/JoaquinCampo/Skills --skill agent-browser-joaquincampo
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/JoaquinCampo/Skills/tree/main/agent-browser
Command: npx skills add https://github.com/JoaquinCampo/Skills --skill agent-browser-joaquincampo

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Browser automation for AI agents to interact with websites, enabling tasks like navigation, form filling, data extraction, visual testing, and automated QA without manual browser control.

Core Features & Use Cases

  • Navigate, snapshot, and interact with page elements via deterministic refs
  • Support for session management, authentication flows, and state persistence
  • Debugging, screenshots, diffs, and data extraction for QA and automation scenarios

Quick Start

Start by opening a URL with agent-browser, then snapshot the page to generate element refs for subsequent automated interactions.

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 tasks like web testing and data extraction with a CLI?

Browser automation for web testing and data extraction is handled by a CLI that controls a headless or headed browser to navigate pages, fill forms, click elements, and scrape data. You start by opening a URL, snapshotting the page, and using generated refs for subsequent interactions.

What is ref-based interaction for headless Chrome automation?

Ref-based interaction in browser automation uses deterministic references generated from page snapshots to identify and interact with elements. This mechanism allows AI agents to reliably click, fill forms, and navigate websites without relying on brittle selectors.

Can I manage authentication workflows and persist sessions in browser automation?

Authentication workflows and session management are supported, allowing you to authenticate sessions across multiple workstreams. The CLI provides session isolation and state persistence so your browser automation can maintain logged-in states across tasks.

Does this browser automation tool support taking screenshots and visual diffs for QA?

Screenshots and visual diffs are supported for debugging and automated QA scenarios. The CLI captures page states during browser automation, enabling you to extract data and perform visual testing without manual browser control.

Do I need a proxy to run headless browser automation for AI agents?

A proxy is optional when running headless browser automation for AI agents. The CLI supports optional proxies alongside session isolation and ref-based interactions, allowing flexible configuration depending on your network and data extraction requirements.

What are the limitations of using a CLI for browser automation?

Limitations of CLI-based browser automation include potential constraints with highly dynamic JavaScript-heavy pages where snapshot refs might shift. It is best suited for deterministic automation tasks like scraping, form filling, and testing rather than complex real-time web app interactions requiring human-like adaptation.