A powerful library that makes websites accessible for AI agents to control browsers and automate complex web operations.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install browser-use-guide
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-guide 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?
browser-use is a state-of-the-art automation framework designed to bridge the gap between Large Language Models and the web. By utilizing Openclaw Skills, developers can empower AI agents to navigate websites, interact with UI elements, and execute multi-step workflows just like a human user. Whether you are using Claude, GPT, Gemini, or local models via Ollama, this skill provides the necessary interface for your computer to 'get things done' on the internet.
The framework is built with a focus on simplicity and flexibility, supporting over 1000+ integrations including Gmail, Slack, and Notion. It offers a unique value proposition by providing both local execution for privacy and cloud-based stealth browsing to bypass sophisticated anti-bot mechanisms and CAPTCHAs, ensuring high reliability for automated tasks.
browser-use Use Cases
- Automatically filling out job applications and uploading resumes.
- Managing e-commerce tasks like adding items to a grocery list or completing orders.
- Conducting price comparison research across multiple retailers for hardware or travel.
- Navigating complex dashboards to extract data or perform administrative tasks.
- Integrating web actions into productivity tools like Slack or Notion via AI agents.
How browser-use Works
- The AI agent receives a natural language task (e.g., 'Find the price of a GPU').
- The Openclaw Skills controller initializes the browser-use Agent and connects it to a specified LLM.
- The Agent analyzes the current browser state, identifying clickable elements and input fields.
- Decisions are made by the LLM to execute specific browser actions such as clicking, typing, or scrolling.
- The process iterates through steps until the goal is achieved or the maximum step limit is reached.
- Screenshots and state data are captured throughout the lifecycle for debugging and verification.
browser-use Setup
Install the core library using uv for modern Python dependency management:
# Initialize environment and add library
uv init && uv add browser-use && uv sync
# Install required Chromium browser
uvx browser-use install
# Integrate with Openclaw Skills
npx clawhub@latest install browser-use
Configure your environment variables in a .env file:
cat > .env << EOF
BROWSER_USE_API_KEY=your-key
ANTHROPIC_API_KEY=your-key
EOF
browser-use Data Schema & Taxonomy
The skill manages its operations through structured classes and configuration files as detailed below:
| Component | Description |
|---|---|
| Agent | The core controller managing the task logic and LLM communication. |
| Browser | Configuration for local Chrome instances or Cloud Stealth sessions. |
| Tools | Custom Python functions registered as executable actions for the AI. |
| State | Metadata including accessibility trees, screenshots, and element indices. |
browser-use Advanced Features
- Cloud Stealth Mode: Use managed browsers with built-in proxy rotation and CAPTCHA solving.
- Custom Action Registration: Define domain-specific Python functions that the AI can call as tools.
- Persistent Profiles: Reuse existing Chrome user profiles to maintain login sessions for Gmail or social media.
- Multi-Model Support: Hot-swap between Anthropic Claude, Google Gemini, or local Ollama instances.
- MCP Integration: Connect as a Model Context Protocol service for seamless use within the Openclaw Skills ecosystem.
SKILL.md
Loading
Related Openclaw Skills
Generate Estate Planning Law Firm Client Education Handout
An automated workflow for estate planning attorneys to generate polished, visual client education materials and FAQ guides.
Executive Recruiting Client Education Handout Generator
Automate the creation of high-quality, visual client education handouts and FAQ guides for executive search firms.
Generate Finance Ad Creative Brief
A specialized AI tool for generating data-driven creative briefs and visual hooks tailored for financial marketing campaigns.
DeepSeek V4
A high-performance CLI tool to interact with DeepSeek V4 Flash and Pro models using an OpenAI-compatible API.
Car Insurance Policy Automator
An automated utility for identifying, renaming, and archiving car insurance policy PDFs using local parsing or AI models.
Logics-Parsing Document Intelligent Analysis
An end-to-end document parsing tool by Alibaba that transforms complex PDFs and images into structured HTML with support for formulas, chemical structures, and code.
Featured*
