A robust HTTP API interface for controlling headless Chrome instances within a Cloudflare Workers environment.
The fastest way to install a skill directly from the registry.
npx clawhub@latest install cloudflare-browser
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
Copy this prompt to OpenClaw to install it automatically.
Help me install cloudflare-browser using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Get the raw skill files in a ZIP archive.
The Cloudflare Browser Rendering skill provides a high-level HTTP interface to manage headless Chrome instances without the complexity of WebSocket CDP connections. Designed to run within Cloudflare Workers, this skill enables developers to perform complex web interactions like full-page screenshots, automated form submission, and dynamic content navigation. By utilizing Openclaw Skills, you can bridge the gap between AI agents and the live web, allowing for real-time data extraction and automated browsing tasks.
This skill is particularly valuable for developers who need reliable browser automation that is both scalable and secure. It abstracts away the infrastructure of managing browser clusters, offering a simple set of endpoints to execute actions like clicking, typing, and running custom JavaScript on any webpage.
To get started with this skill, ensure your Cloudflare environment is configured and set the necessary environment variables:
export MOLTBOT_URL="https://your-worker.workers.dev"
export CDP_SECRET="your-secret"
You must have a Cloudflare Worker deployed with the Browser Rendering binding enabled to use these Openclaw Skills effectively.
The skill organizes its data through structured JSON payloads for both requests and responses. Key components include:
| Data Component | Description |
|---|---|
| Viewport | Custom width and height settings for the browser window. |
| Actions | An array of steps including click, type, wait, and screenshot. |
| Results | Extracted text content, page titles, and final URLs after navigation. |
| Screenshots | Base64-encoded image data in PNG or JPEG format. |
| Metadata | Information on execution time, success status, and error logs. |
waitForSelector to handle asynchronous page loads and dynamic elements.Loading
A utility that converts CHANGELOG.md files into RSS 2.0 feeds to enable automated project release monitoring.

A structural comparison tool for analyzing and merging YAML, JSON, TOML, and INI configuration files.

An AI-powered automation engine designed to connect enterprise tools, create complex pipelines, and eliminate manual business tasks.

A command-line driven utility for analyzing, cleaning, and organizing large browser bookmark export files.

A comprehensive automation skill for managing the end-to-end sales lifecycle on the Coconala marketplace using AI-driven browser interactions.

Automatically transform your codebase into comprehensive API references and user guides with minimal manual effort.








































