agent-browser

Automate browser navigation, element interaction, form filling, and screenshots.

Updated Mar 24, 2026
One-click install
npx skills add https://github.com/1thirteeng3/greenmoire --skill agent-browser-1thirteeng3
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/1thirteeng3/greenmoire/tree/main/integrations/oh-my-openagent/src/features/builtin-skills/agent-browser
Command: npx skills add https://github.com/1thirteeng3/greenmoire --skill agent-browser-1thirteeng3

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The agent-browser skill automates routine browser tasks for web testing, form filling, screenshots, and data extraction, reducing manual effort and speeding up validation workflows.

Core Features & Use Cases

  • Navigate to pages and manage sessions
  • Snapshot interactive elements and interact via element refs
  • Take screenshots, generate reports, and extract data from web pages
  • Use advanced probing like semantic locators and CDP-backed actions for modern apps

Quick Start

Open a URL with agent-browser, snapshot interactive elements to obtain refs, and interact using those refs.

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 testing and extract data from web pages?

Automate browser testing by scripting navigation, form filling, and element interaction using deterministic commands. You snapshot interactive elements to obtain refs, interact via those refs, and capture screenshots to validate modern web apps.

How does CDP support work for headless browser automation commands?

CDP-backed actions drive headless browser automation by executing advanced probing and semantic locators. This enables scripted workflows with deterministic commands to interact with modern web apps without a visible UI.

What is the best way to fill forms and take screenshots during web testing?

Use scripted workflows with deterministic commands to fill forms and take screenshots during web testing. Navigate to a page, snapshot interactive elements to get refs, fill forms via those refs, and capture screenshots for validation.

Can I use semantic locators to interact with elements in modern web apps?

Yes, you can use semantic locators and CDP-backed actions to interact with elements in modern web apps. The skill snapshots interactive elements to obtain refs, enabling you to interact with them using deterministic commands.

Does this browser automation tool support both headless and headed operation modes?

Yes, this browser automation tool supports optional headless and headed operation modes. It covers navigation, element interaction, form filling, and screenshots across modern web apps using deterministic commands for flexible scripted workflows.

Why should I use element refs for web testing instead of direct selectors?

Using element refs for web testing stabilizes interactions by snapshotting interactive elements first. You then interact using deterministic refs, reducing flakiness from dynamic DOM changes in modern web apps during data extraction.