agent-browser

Automate web page interactions via agent-browser CLI with ref-based element selection.

Updated Feb 25, 2026
One-click install
npx skills add https://github.com/robwestz/bacowr_b --skill agent-browser-robwestz
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/robwestz/bacowr_b/tree/main/every-marketplace/plugins/compound-engineering/skills/agent-browser
Command: npx skills add https://github.com/robwestz/bacowr_b --skill agent-browser-robwestz

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automate web interactions for AI agents via a CLI, enabling hands-free browsing, form filling, and data collection.

Core Features & Use Cases

  • Full CLI-based browser control with ref-based element targeting for deterministic interactions.
  • Snapshot, open, click, fill, type, get, and scrape capabilities across websites.
  • Use cases include automated testing, data extraction, form submission, and visual validation via screenshots.

Quick Start

Open a URL with agent-browser open, snapshot interactive elements, then interact using refs to complete tasks.

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 scraping using a CLI for AI agents?

Automate web scraping by using agent-browser's CLI to open URLs, snapshot interactive elements, and extract data. It enables AI-assisted workflows to collect information across various websites through scripted commands without manual browsing.

How does ref-based element selection work for browser automation?

Ref-based element selection identifies interactive web page elements through snapshot references, enabling deterministic interactions like clicking and filling. This ensures AI agents target the exact UI components reliably during automated testing or form submission workflows.

Can I perform automation testing across websites with agent-browser?

Yes, agent-browser supports automation testing by executing scripted CLI commands for deterministic operations. You can fill forms, click elements, and take screenshots to validate visual output across various websites.

What is the best way to automate form filling via headless browser commands?

Automate form filling by using CLI commands to open a URL, snapshot the page, and use refs to target input fields. This allows AI agents to script type and fill actions for hands-free form submission.

Do I need to install agent-browser to use scripted web interactions?

Yes, installation of agent-browser is required. Once installed, you can use its CLI to execute scripted commands, take snapshots, and perform deterministic web interactions for your AI-assisted workflows.