OpenClaw Browser for Openclaw

A powerful browser automation skill for OpenClaw using the Chrome DevTools Protocol to capture screenshots and interact with web pages.

roger0808
v1.0.0
Feb 15, 2026
0
0
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install openclaw-browser

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install openclaw-browser 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 OpenClaw Browser?

The OpenClaw Browser skill enables seamless browser automation by leveraging the Chrome DevTools Protocol (CDP). It allows AI agents to interact with live web pages, perform automated tasks, and capture high-quality screenshots. By connecting to a running instance of Chrome, this skill bypasses common headless detection mechanisms found on complex websites, making it a reliable choice for robust web data extraction and visual verification within the Openclaw Skills ecosystem.

This skill is designed for developers who need more than just simple HTML parsing. By utilizing a real browser instance, it handles JavaScript rendering, dynamic content, and complex user interactions that traditional scrapers often miss.

OpenClaw Browser Use Cases

  • Capturing full-page screenshots of websites for documentation or monitoring.
  • Automating interactions with web applications that require a real browser environment.
  • Bypassing anti-bot and headless browser detection on modern web platforms like Taobao or Xiaohongshu.
  • Visual regression testing and UI verification during the development lifecycle.

How OpenClaw Browser Works

  1. Establishes a connection to a running Chrome instance via the Chrome DevTools Protocol at port 9222.
  2. Initializes a new browser tab or selects an existing session for the task.
  3. Navigates to the specified target URL and waits for the page lifecycle to complete.
  4. Executes automated actions or triggers a screenshot capture of the rendered content.
  5. Saves the output file to a designated local path for further processing or agent feedback.

OpenClaw Browser Setup

First, ensure Chrome is installed and start it with remote debugging enabled using the following command:

chrome --remote-debugging-port=9222 --remote-debugging-address=0.0.0.0

To capture a screenshot using your Openclaw Skills setup, run the provided Node.js script:

node ~/.openclaw/workspace/skills/openclaw-browser/scripts/screenshot.js https://example.com /tmp/output.png

OpenClaw Browser Data Schema & Taxonomy

The skill manages browser sessions and output files through the following schema:

Component Detail
Connection URL http://127.0.0.1:9222
Script Location scripts/screenshot.js
Supported Formats PNG, JPEG (via filesystem output)
Customization Supports --width and --height flags for viewport control

OpenClaw Browser Advanced Features

  • Full-page capture capability that automatically scrolls to include all content beyond the initial viewport.
  • Custom viewport configuration for testing responsive designs and mobile layouts.
  • Direct Puppeteer integration support for creating complex, multi-step automation scripts and interactions.
  • Non-headless execution mode to remain undetectable by advanced bot-mitigation services.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*