An AI-powered browser automation skill that enables natural language interaction with websites through a high-level CLI.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install openclaw-browser-stagehand
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 openclaw-browser-stagehand 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 Automation?
The Browser Automation skill is a robust tool designed for Openclaw Skills that allows AI agents to navigate the web, perform complex actions, and extract data using simple natural language commands. By leveraging the Stagehand CLI, it removes the need for complex CSS selectors or XPaths, allowing developers to automate browser tasks as if they were instructing a human. It provides a flexible architecture that supports both local development and production-grade scraping environments.
Whether you are building a research agent or a web-testing bot, this skill provides the necessary bridge between LLM reasoning and the live web. It integrates seamlessly into the Openclaw Skills ecosystem, providing a standardized way to handle web-based workflows without manually writing repetitive automation scripts.
Browser Automation Use Cases
- Automating data extraction from complex web applications using natural language queries.
- Performing end-to-end testing of web workflows like signing in, form filling, and navigation.
- Researching information across multiple websites and capturing screenshots for verification.
- Building autonomous agents that can interact with SaaS tools and internal web dashboards.
How Browser Automation Works
- Environment Detection: The skill automatically detects if Browserbase API keys are present in the .env file to choose between local Chrome or remote execution.
- Navigation: The agent initiates a session and navigates to a specified URL using the navigate command.
- Natural Language Actions: Users provide intent-based instructions (e.g., "click the login button") which are translated into browser interactions.
- Data Extraction: Structured data is retrieved from the page based on natural language schemas or instructions.
- Verification: Screenshots can be taken at any stage to allow the AI agent to verify the current state of the UI.
- Session Cleanup: The browser instance is closed efficiently once the workflow is completed.
Browser Automation Setup
To begin using this skill within the Openclaw Skills framework, follow these installation steps in the skill directory:
npm install
npm link
If you require stealth mode or CAPTCHA solving, ensure your .env file contains your BROWSERBASE_API_KEY and BROWSERBASE_PROJECT_ID.
Browser Automation Data Schema & Taxonomy
The Browser Automation skill handles various data types to facilitate seamless web interaction:
| Component | Type | Description |
|---|---|---|
| Navigation URL | String | The target web address for the session. |
| Extraction Schema | JSON String | Optional structure defining how data should be formatted upon extraction. |
| Screenshots | Image/File | Visual output used for debugging and verification. |
| Observation Data | List | A collection of discovered interactive elements identified by the CLI. |
Browser Automation Advanced Features
- Intelligent environment switching between Local Chrome for speed and Browserbase for production reliability.
- Built-in stealth mode and proxy support when running in remote environments.
- Natural language element discovery using the observe command to handle dynamic UIs.
- Seamless integration with other Openclaw Skills via standardized CLI commands.
SKILL.md
Loading
Related Openclaw Skills
Central Key Management System
An enterprise-grade platform for the secure generation, verification, and centralized lifecycle management of cryptographic API keys.
Shell Image & Video Skill
A powerful AI media generation toolkit for face swapping, motion transfer, and digital human creation via RunningHub AI.
Incident Response Playbook Generator API
Automate the creation of tailored incident response playbooks based on organizational data and security compliance requirements.
ClawInboxRAG
A secure, read-only skill for parsing and executing mailbox retrieval commands against a local Gmail RAG backend.
TOC Stock AI Trading Assistant
An intelligent stock analysis and simulation system designed to provide AI-driven market insights and portfolio management.
Skill Quality Check
A universal quality assessment framework designed to audit, score, and optimize SKILL.md files for any AI agent environment.
Featured*
