A robust browser automation CLI designed for AI agents to navigate, interact with, and extract data from web pages using persistent sessions.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install agent-browser-4
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install agent-browser-4 using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Agent Browser?
Agent Browser is a specialized Command Line Interface (CLI) that enables AI agents to perform complex web automation tasks with ease. By providing a structured way to interact with the Document Object Model (DOM) through unique element references, it bridges the gap between raw web code and agentic reasoning. This tool is essential for developers building Openclaw Skills that require reliable web navigation, form filling, and real-time data extraction.
The tool excels at handling the dynamic nature of modern web applications by offering features like session persistence, automatic element discovery, and visual debugging. It allows Openclaw Skills to maintain state across multiple commands, making it ideal for tasks involving multi-page workflows or authenticated sessions.
Agent Browser Use Cases
- Automating multi-step form submissions and user registrations.
- Scraping dynamic or JavaScript-heavy content from modern websites.
- Performing automated end-to-end testing for web applications.
- Managing authenticated sessions to perform tasks behind login screens.
- Generating visual reports using annotated screenshots and PDF exports.
- Simulating mobile browser environments for cross-platform testing.
How Agent Browser Works
- Initialize a browser session by opening a target URL using the open command.
- Generate a snapshot of the page to identify interactive elements and assign unique references (e.g., @e1, @e2).
- Execute interactions such as clicking, typing, or selecting options using the identified element refs.
- Use wait commands to ensure network activity is idle or specific DOM elements have rendered.
- Re-snapshot the page after any navigation or state change to refresh the reference pointers.
- Optionally save the session state to a JSON file to persist cookies and authentication for future interactions.
Agent Browser Setup
To get started with this tool in your Openclaw Skills environment, install the package and its dependencies:
# Install the agent-browser package
npm install -g agent-browser
# (Optional) For iOS simulation support
npm install -g appium
appium driver install xcuitest
# Run a simple test command
agent-browser open https://example.com && agent-browser snapshot -i
Agent Browser Data Schema & Taxonomy
Agent Browser organizes its data and configuration using structured formats to ensure compatibility with various Openclaw Skills workflows:
| Component | Format | Description |
|---|---|---|
| Session State | JSON | Contains cookies, localStorage, and session metadata for persistence. |
| Element Snapshot | JSON/Text | Maps interactive elements to @ref tags based on accessibility roles. |
| Configuration | JSON | Stores global or project-specific settings like headless mode and proxy info. |
| Captured Media | PNG/PDF/WebM | Stores screenshots, document exports, and session recordings. |
Agent Browser Advanced Features
- Session Persistence: Seamlessly save and restore browser states, including cookies and localStorage, to bypass repetitive login steps.
- Annotated Screenshots: Generate 'Vision Mode' captures where interactive elements are visually labeled with their corresponding references.
- Parallel Session Management: Run and track multiple isolated browser instances simultaneously using named sessions.
- JavaScript Evaluation: Execute complex scripts in the browser context with support for stdin and base64 encoding to avoid shell escaping issues.
- iOS Simulator Integration: Automate mobile Safari on macOS using Xcode and Appium for mobile-specific workflows.
- Connection to Existing Chrome: Attach the CLI to a running Chrome instance via the Chrome DevTools Protocol (CDP).
SKILL.md
Loading
Related Openclaw Skills
YT-to-Blog Content Engine
A comprehensive AI-powered pipeline that converts a single YouTube URL into a complete cross-platform content suite including blogs, social threads, and AI avatar videos.
GEDCOM Explorer
A Python-based tool that transforms static GEDCOM files into interactive, single-file HTML genealogy dashboards.
Dory-Memory
A robust file-based memory framework that ensures AI agents maintain task continuity and context across session resets.
RSSaurus CLI
A high-performance Go binary that enables full interaction with rssaurus.com via the terminal for efficient RSS feed triage and automation.
Persistent Code Terminal
A persistent tmux-based session manager designed to provide AI agents with stable, long-running terminal environments for complex coding tasks.
Elsa x402 DeFi Tools
An advanced DeFi toolkit for portfolio management and automated swap execution powered by x402 micropayments.
Featured*
