agent-browser

Automate browser tasks via CLI with navigate, click, fill, and screenshot commands.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/bketelsen/ytsejam --skill agent-browser-bketelsen
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/bketelsen/ytsejam/tree/main/contrib/skills/agent-browser
Command: npx skills add https://github.com/bketelsen/ytsejam --skill agent-browser-bketelsen

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

The agent-browser skill eliminates the manual, repetitive burden of controlling a browser to perform web tasks so AI agents can automate page navigation, data extraction, form interaction, and screenshot capture with consistent behavior.

Core Features & Use Cases

  • Deterministic browser automation via a CLI that supports navigate, snapshot, click, fill, type, and screenshot, with compact refs for AI reasoning.
  • Multi-session isolation and iframe-aware inlining to enable concurrent workflows and robust site interactions.
  • Use cases include logging in to sites, extracting data from pages, testing web apps, and automating routine browser tasks in AI-assisted workflows.

Quick Start

Install and start using agent-browser to automate a site by opening a page and beginning 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 with an AI agent?

Browser tasks are automated for AI agents using a CLI to navigate, click, fill, and screenshot pages while exposing compact refs for efficient reasoning. This enables consistent page navigation, data extraction, and form interaction within AI-assisted workflows.

Does AI agent browser automation require a specific web browser?

AI agent browser automation requires a Chromium-based browser to function. It leverages this environment to drive web page interactions, manage multi-session isolation, and support optional references, scripts, and assets directories to extend capabilities.

How do I extract data from web pages using an AI agent?

Data extraction from web pages is handled by navigating to the target URL and using snapshot functionality with compact refs. These refs expose page structure for AI reasoning, enabling agents to locate and extract specific data efficiently.

Can I run multiple browser automation sessions concurrently for AI workflows?

Multi-session isolation enables concurrent browser automation workflows for AI agents. This feature supports running multiple isolated sessions simultaneously, allowing robust site interactions, tab management, and independent task execution without conflicts.

What is the best way to handle iframe content during web automation?

Iframe content is handled through iframe-aware inlining during web automation. This mechanism integrates iframe content into the page snapshot, allowing AI agents to reason about and interact with elements inside iframes seamlessly.

Why use compact refs for AI reasoning in browser automation?

Compact refs are used in browser automation to provide AI agents with efficient, deterministic references to page elements. This reduces the reasoning overhead by exposing targeted interaction points instead of raw DOM trees for tasks like clicking and filling forms.