agent-browser

agent-browser enables autonomous web automation by letting AI agents interact directly with browser elements. Control clicks, form submissions, page navigation, and DOM inspection through a unified interface designed for seamless integration with Claude and other AI platforms.

agent-browser is a web browser automation tool that enables AI agents to interact directly with browser elements. It allows you to control clicks, form submissions, page navigation, and DOM inspection through a unified interface designed for seamless integration with Claude and other AI platforms. The tool is released under the MIT license, making it freely available for both commercial and personal use.

AI-generated summary based on this skill's SKILL.md

★ 184 ⁮ 60 MIT updated 2026-07-26 by sbusso

Install

sbusso/claudeclaw/agent-browser · repository language: TypeScript

Manual (git clone) CLI (skillfed)coming soon

git clone https://github.com/sbusso/claudeclaw
cp -r claudeclaw/agent/skills/agent-browser ~/.claude/skills/agent-browser

Frequently asked questions

AI-generated answers based on this skill's SKILL.md and metadata

What is agent-browser and what can it do?

How do I use agent-browser to automate clicking and filling forms?

agent-browser provides programmatic control over browser interactions, allowing you to automate clicking elements and filling forms on web pages. You can instruct the tool to locate form fields, input data, and submit forms without manual intervention. This automation capability is one of agent-browser's primary strengths, enabling you to streamline repetitive web tasks and integrate them into larger automation workflows.

Can agent-browser extract data and content from websites?

Yes, agent-browser supports web scraping and data extraction by allowing you to programmatically access website content. You can inspect the DOM, retrieve text and attributes from page elements, and extract structured data from web pages. This makes agent-browser useful for gathering information from websites automatically without manual copying and pasting.

What is agent-browser's primary use case for web application testing?

agent-browser excels at test web applications and capture screenshots for verification purposes. You can automate browser testing workflows by navigating through application flows, performing user actions, and taking screenshots at key points to verify application behavior. This capability makes agent-browser valuable for quality assurance and automated testing scenarios.

How does agent-browser integrate with AI platforms like Claude?

agent-browser is specifically designed for seamless integration with Claude and other AI platforms. The unified interface allows AI agents to autonomously interact with web pages, making decisions about which elements to click, what data to extract, and how to navigate. This integration enables AI systems to perform complex web automation tasks that would otherwise require manual scripting or human intervention.

Can agent-browser manage browser state like cookies and authentication?

agent-browser provides capabilities to manage browser state including authentication, cookies, and storage. This allows you to maintain persistent sessions across multiple interactions, handle authenticated workflows, and preserve browser context between operations. State management is essential for automating complex web application interactions that depend on user sessions and stored data.

SKILL.md

rendered from the published skill — quoted content, verbatim

Browser Automation with agent-browser

Quick start

agent-browser open &lt;url&gt;        # Navigate to page
agent-browser snapshot -i       # Get interactive elements with refs
agent-browser click @e1         # Click element by ref
agent-browser fill @e2 "text"   # Fill input by ref
agent-browser close             # Close browser

Core workflow

Navigate: agent-browser open <url>
Snapshot: agent-browser snapshot -i (returns elements with refs like @e1, @e2)
Interact using refs from the snapshot
Re-snapshot after navigation or significant DOM changes

Commands

Navigation

agent-browser open &lt;url&gt;      # Navigate to URL
agent-browser back            # Go back
agent-browser forward         # Go forward
agent-browser reload          # Reload page
agent-browser close           # Close browser

Snapshot (page

(truncated - see the full file via the links below)

Read as markdown · JSON record · Browse the source repository

File tree — 1 file

agent/skills/agent-browser/SKILL.md

All skills by sbusso → Best browser automation skills for AI agents - SkillFed →