A powerful CLI skill for automating web browser interactions and data extraction using natural language commands.
The fastest way to install a skill directly from the registry.
npx clawhub@latest install agent-browser-stagehand
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 agent-browser-stagehand using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Get the raw skill files in a ZIP archive.
The Browser Automation skill leverages the Stagehand CLI to bridge the gap between AI reasoning and web navigation. By integrating this capability into Openclaw Skills, developers can perform complex web tasks like clicking buttons, filling forms, and extracting structured data without writing custom Playwright scripts. The skill is designed to handle both local development and production-scale scraping through its flexible environment selection.
This tool enables AI agents to interact with the live web just as a human would, but with the speed and precision of a machine. It supports a variety of commands that translate simple instructions into executable browser actions, making it an essential component for any workflow requiring real-time web data or automated web application testing.
To get started with this skill within the Openclaw Skills ecosystem, follow these installation steps:
# Navigate to the skill directory
# Install the necessary Node.js dependencies
npm install
# Link the browser command globally for CLI access
npm link
To enable remote execution, ensure your .env file contains your BROWSERBASE_API_KEY and BROWSERBASE_PROJECT_ID.
The skill manages browser state and data extraction through structured CLI outputs. Key data components include:
| Component | Format | Description |
|---|---|---|
| URL | String | The target web address for navigation |
| Action | String | Natural language instruction for UI interaction |
| Extraction Schema | JSON | Optional structure defining how data should be returned |
| Screenshot | File | PNG image of the current browser viewport |
| Observation Map | List | A discovery of interactable elements on the page |
Loading
An AI-optimized web search skill that delivers clean, relevant web data and deep research capabilities for Openclaw Skills agents.

A unified configuration suite to connect OpenClaw AI agents with popular messaging platforms like Telegram, Slack, and Discord.

An autonomous framework for running multi-agent AI trading competitions with real-time market data and evolutionary strategy selection.

A 24/7 autonomous AI trading system where five agents compete in a daily tournament on OKX to refine strategies.

An AI-powered browser automation tool that executes natural language commands to navigate, interact with, and extract data from any website.

A natural language interface for automating browser actions, web navigation, and data extraction via CLI.








































