A comprehensive web automation skill for OpenClaw that enables professional headless browser interaction, data extraction, and visual page capture.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install browserless-agent
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 browserless-agent 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 Browserless Agent?
The Browserless Agent is a high-performance web automation skill designed to extend the capabilities of the OpenClaw ecosystem. It provides a robust interface for interacting with any website using a headless browser, eliminating the need for local browser installations. By connecting to a remote or local Browserless instance, this skill allows users to perform complex web tasks programmatically, making it a cornerstone of the Openclaw Skills library for developers.
This skill is built for reliability and speed, offering deep control over page navigation, DOM manipulation, and network requests. Whether you are building an automated research tool or a web-based testing suite, the Browserless Agent provides the necessary infrastructure to interact with modern, dynamic web applications seamlessly.
Browserless Agent Use Cases
- Extracting structured data and text from dynamic, JavaScript-heavy websites.
- Generating high-resolution screenshots and custom PDF reports for web pages.
- Automating complex multi-step workflows like form filling, logins, and file uploads.
- Performing automated UI testing and verifying element states across different device emulations.
- Researching web content by navigating iFrames and handling browser storage like cookies and localStorage.
How Browserless Agent Works
- The user provides a command or URL to the AI agent which then selects the appropriate action from the Browserless Agent.
- The skill initiates a WebSocket connection to the configured Browserless service using the Playwright protocol.
- The browser instance performs the requested lifecycle actions, such as navigating to a URL or waiting for specific DOM elements to load.
- Advanced logic handles dynamic content by waiting for network idle states or executing custom JavaScript within the page context.
- The results, including extracted data or generated files, are processed and returned to the Openclaw Skills environment for further use.
Browserless Agent Setup
To integrate this skill, you must configure your environment variables within your OpenClaw settings.
- Locate the browserless-agent in your Skills settings.
- Set the mandatory
BROWSERLESS_URLand optionalBROWSERLESS_TOKENif using a cloud provider.
# Cloud Service Configuration
BROWSERLESS_URL=wss://chrome.browserless.io
BROWSERLESS_TOKEN=your-private-token
# Local Service Configuration
BROWSERLESS_URL=ws://localhost:3000
The skill automatically appends the necessary Playwright endpoints and authentication parameters based on your Openclaw Skills configuration.
Browserless Agent Data Schema & Taxonomy
The Browserless Agent organizes data through structured JSON payloads for actions and standardized outputs for results:
| Data Type | Description | Key Fields |
|---|---|---|
| Navigation | Controls page state | url, timeout, wait_until |
| Extraction | Retrieves DOM data | selector, attribute, all, type |
| Interaction | Simulates user input | text, delay, clear, force |
| Visuals | File generation | path, quality, full_page, format |
| Network | Request management | headers, resource_types |
Files generated (like screenshots and PDFs) are saved to the local file system at the path specified in the action payload.
Browserless Agent Advanced Features
- Support for device emulation including iPhone and Galaxy profiles to test responsive layouts.
- Resource blocking capabilities to save bandwidth by preventing images, fonts, or stylesheets from loading.
- Comprehensive storage management for interacting with cookies, localStorage, and sessionStorage.
- Multi-tab support allowing the agent to manage and switch between several open pages simultaneously.
- Geolocation spoofing to simulate browsing from specific geographical coordinates.
SKILL.md
Loading
Related Openclaw Skills
wacli
A specialized command-line interface for sending WhatsApp messages to third parties and managing message history.
Signet Guardian
A policy-driven firewall for AI agents that enforces spending limits and merchant restrictions before payments occur.
Flight Tracker
Track real-time flight details, including status, gates, and live positions, with a professional Flighty-style visual output.
Organizze
A comprehensive tool for managing personal finances, from tracking bank transactions to monitoring budget goals, directly through an AI agent.
ClawSaver
ClawSaver is an intelligent request-batching utility that reduces AI model API costs by 20–40% through efficient message buffering.
ClawText Ingest
ClawText Ingest provides production-ready memory ingestion for AI agents, supporting Discord, files, and URLs with automatic deduplication.
Featured*
