skills.rest
drhazemibclc
drhazemibclcCOMMUNITY
Explore all skills by this author →
GitHub

agent-browser

Drive a headless browser via CLI using ref-based element selection.

Updated Mar 22, 2026
Search Tags:#automation#cli#screenshot#refs#web-automation#agent-browser
One-click install
npx skills add https://github.com/drhazemibclc/plate --skill agent-browser-drhazemibclc
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/drhazemibclc/plate/tree/main/.codex/skills/agent-browser
Command: npx skills add https://github.com/drhazemibclc/plate --skill agent-browser-drhazemibclc

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automates browser interactions for AI agents by exposing a CLI to drive a headless browser using ref-based element selection.

Core Features & Use Cases

  • Headless browser automation via CLI for AI agents.
  • Ref-based element selection using accessibility snapshots (@e1, @e2).
  • Navigation, interaction, screenshot capture, and data extraction across web pages.
  • Use cases include automated login flows, form submissions, data scraping, and website testing.

Quick Start

Open a URL, snapshot the accessibility tree, and interact with elements using refs to automate a task.

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 tasks using a headless browser via CLI?▼

Web task automation via CLI is handled by driving a headless browser using ref-based element selection, enabling navigation, form filling, and data scraping. It requires installing the agent-browser tool globally and downloading a Chromium instance.

What is ref-based element selection for web automation?▼

Ref-based element selection for web automation uses accessibility snapshots to reference page elements like @e1 or @e2. This method allows AI agents to interact with specific elements without relying on traditional CSS selectors or XPath queries.

Can I take screenshots and scrape data from web pages using CLI automation?▼

Screenshots and data scraping are supported through CLI automation by navigating pages and capturing the output. The headless browser executes these actions across sites to validate web flows or extract required information.

Do I need to install Chromium separately to use headless browser automation?▼

Chromium installation is required for headless browser automation, initiated via the command 'agent-browser install' after globally installing the tool. This downloads the necessary browser instance to execute automated web interactions.

Does this approach work for automating login flows and form submissions?▼

Automating login flows and form submissions is fully supported by navigating to the target URL, snapshotting the accessibility tree, and interacting with input elements using their assigned refs to complete the required actions.

What are the limitations of using accessibility snapshots for element selection?▼

Accessibility snapshot limitations include dependency on the browser's rendered accessibility tree, meaning elements not exposed in the tree cannot be selected using refs like @e1, restricting interaction with dynamically obscured components.

Related Skills

Explore More Software Engineering Skills→
SomtoUgehSomtoUgeh

agent-browser

Automate browser interactions via agent-browser CLI with ref-based selectors.

Community
Intermediate
landonking-giflandonking-gif

deep_research

Search the web, extract data points, and prepare structured report content.

Community
Advanced
xingxingstars123xingxingstars123

agent-browser

Automate browser interactions with ref-based element targeting and JSON snapshots.

Community
Advanced
Mte90Mte90
51

core-actionbook

Retrieve structured browser automation action manuals by ID and keyword.

Community
Intermediate