ClawKit Logo
ClawKitReliability Toolkit
Back to Registry
Official Verified browser automation Safety 3/5

browser

Automate web browser interactions using natural language via CLI commands. Use when the user asks to browse websites, navigate web pages, extract data from websites, take screenshots, fill forms, click buttons, or interact with web applications.

Why use this skill?

Automate web browsing, data extraction, and form filling with the OpenClaw browser skill. Supports local and remote headless environments.

skill-install — Terminal

Install via CLI (Recommended)

clawhub install openclaw/skills/skills/peytoncasper/agent-browser-stagehand
Or

What This Skill Does

The Browser skill acts as a powerful interface for automating complex web interactions using the Stagehand CLI engine coupled with Claude's reasoning capabilities. It abstracts the complexity of headless browser automation, allowing users to issue natural language commands instead of writing boilerplate code. Whether you need to navigate to a site, interact with DOM elements, or perform complex data extraction, this skill handles the heavy lifting. It intelligently toggles between local Chrome instances for rapid development and remote Browserbase environments for production-grade scraping, stealth, and proxy support.

Installation

To add the browser skill to your agent, run the install command in your project terminal: clawhub install openclaw/skills/skills/peytoncasper/agent-browser-stagehand. Once the installation is complete, verify the setup status by inspecting the setup.json file in the skill directory. If the setup is not marked as complete, navigate to the skill directory and run npm install followed by npm link to register the global browser command. If you intend to use the remote Browserbase features, ensure your environment variables (BROWSERBASE_API_KEY and BROWSERBASE_PROJECT_ID) are correctly configured in your .env file.

Use Cases

This skill is designed for scenarios requiring programmatic web interaction. Common use cases include: automated web testing to ensure UI components are functional; data scraping where unstructured web content needs to be converted into structured JSON formats; automated form filling for repetitive administrative tasks; and monitoring web applications for specific changes or status updates. It is particularly effective for workflows that require bypassing basic anti-bot protections via Browserbase.

Example Prompts

  1. "Open the browser, navigate to the login page for my dashboard, and enter my credentials."
  2. "Go to the documentation site and extract all the h2 tags into a structured JSON list."
  3. "Browse to the news website and take a screenshot of the main header to verify it loaded correctly."

Tips & Limitations

For the best results, always prioritize navigation commands before attempting interactions. If an action fails, use the browser observe command to see the current state of the DOM; this is invaluable for debugging element selectors. Remember that local execution is faster but lacks the advanced stealth features of remote environments. Always close your browser sessions to prevent resource leakage, especially when running multiple tasks in a sequence.

Metadata

Stars1217
Views2
Updated2026-02-20
View Author Profile
AI Skill Finder

Not sure this is the right skill?

Describe what you want to build — we'll match you to the best skill from 16,000+ options.

Find the right skill
Add to Configuration

Paste this into your clawhub.json to enable this plugin.

{
  "plugins": {
    "official-peytoncasper-agent-browser-stagehand": {
      "enabled": true,
      "auto_update": true
    }
  }
}

Tags(AI)

#browser#automation#web-scraping#headless#testing
Safety Score: 3/5

Flags: network-access, external-api, code-execution