A powerful AI-driven browser automation skill that handles complex multi-step web workflows, logins, and anti-bot measures using the Browser-Use framework.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install browser-use-pro
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 browser-use-pro 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 Browser-Use AI Automation?
Browser-Use is a sophisticated automation skill designed for Openclaw Skills that leverages large language models to navigate the web like a human. Unlike basic scrapers, it can interpret visual cues, handle complex login sequences, and navigate sites protected by anti-bot measures. By integrating the Browser-Use framework with Playwright, this skill allows AI agents to execute reliable sequences of five or more steps, such as navigating to a specific page, filling out forms, and submitting data with reasoning-based error recovery.
Built with a focus on security and privacy, the skill ensures that sensitive credentials never reach the LLM directly through the use of placeholder tokens. It supports both a built-in Chromium instance for speed and a real Chrome CDP connection for scenarios where a user's active session or higher stealth is required. This makes it an essential tool for developers needing robust web interaction capabilities within their Openclaw Skills workflows.
Browser-Use AI Automation Use Cases
- Automating complex multi-step workflows like logging into a portal to submit a support ticket.
- Bypassing sophisticated anti-bot protections on websites using real Chrome instances.
- Batch processing repetitive web operations that require human-like reasoning and navigation.
- Interacting with dynamic web applications where standard CSS selectors or XPaths frequently change.
- Performing automated testing or data entry tasks that require visual confirmation of page states.
How Browser-Use AI Automation Works
- The environment is checked for the existence of the browser-use-env virtual environment to ensure all dependencies are ready.
- The system selects between Mode A (built-in Chromium) for speed or Mode B (Real Chrome CDP) for handling authenticated sessions or bot detection.
- A Python script is dynamically generated based on the user's task, incorporating specific steps and sensitive data handling.
- The script executes locally using the Browser-Use agent, which calls an LLM to reason through each action based on screenshots and DOM state.
- Upon completion, the agent reports the final outcome or enters a troubleshooting loop if roadblocks like CAPTCHAs are encountered.
Browser-Use AI Automation Setup
First, create a virtual environment and install the necessary dependencies for this Openclaw Skills extension:
python3 -m venv ~/browser-use-env
source ~/browser-use-env/bin/activate
pip install browser-use playwright langchain-openai
playwright install chromium
To use Mode B (Real Chrome), launch Chrome with the debugging port enabled:
/Applications/Google\ Chrome.app/Contents/MacOS/Google\ Chrome --remote-debugging-port=9222 &
Browser-Use AI Automation Data Schema & Taxonomy
The skill manages its environment and execution data using the following structure to maintain clean Openclaw Skills operations:
| Component | Path / Format | Description |
|---|---|---|
| Environment | ~/browser-use-env |
Python virtual environment containing all libraries. |
| Task Profile | ~/.browser-use/task-profile |
Local storage for session data and cookies in Mode A. |
| Scripts | .py files |
Dynamically generated Python scripts containing the agent logic. |
| LLM Config | Environment Variables | API keys and base URLs for OpenAI-compatible providers. |
| Sensitive Data | sensitive_data dict |
Temporary mapping of placeholders to real credentials during runtime. |
Browser-Use AI Automation Advanced Features
- CDP Mode for connecting to an existing, authenticated Google Chrome session for maximum stealth.
- Vision-based navigation allowing the AI to see and interact with visual elements instead of just the DOM.
- Domain restriction via the
allowed_domainsparameter to ensure agent safety and compliance within the Openclaw Skills ecosystem. - Fallback LLM support to maintain reliability if the primary model encounters rate limits or errors.
- Sensitive data masking that prevents actual passwords from being sent to the LLM reasoning engine.
SKILL.md
Loading
Related Openclaw Skills
Strava Python Integration
A Python-powered skill for fetching Strava activity data and workout statistics through your AI agent.
Product Hunt Launch
A CLI-driven tracking tool for monitoring Product Hunt launch metrics and leaderboards in real-time.
PostSyncer
A specialized tool for managing social media workspaces and automating post scheduling through AI-driven workflows.
Lemon Squeezy Admin
An administrative command-line interface for managing Lemon Squeezy storefronts, orders, and subscription metrics.
Codex Review
A comprehensive three-tier code quality defense and security audit orchestration layer for users of Openclaw Skills.
Debug Methodology
A rigorous, five-phase framework for AI agents to perform systematic root cause analysis and safe code deployment.
Featured*
