A high-performance browser automation skill that enables AI agents to control web browsers, navigate sites, and interact with elements via the Model Context Protocol.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install playwright-mcp-1-0-0
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 playwright-mcp-1-0-0 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 Playwright MCP?
The Playwright MCP skill is a comprehensive automation engine designed to give AI agents full control over web browsers like Chromium, Firefox, and WebKit. By leveraging the industry-standard Playwright framework, this addition to Openclaw Skills allows for seamless programmatic interaction with any website, handling complex tasks like JavaScript rendering, iframe interaction, and session management.
Whether you are building an agent for automated testing, data mining, or web-based task execution, this skill provides the necessary tools to navigate URLs, click buttons, fill out forms, and extract structured information from the DOM. It functions as a bridge between the Model Context Protocol (MCP) and modern web engines, ensuring high reliability and speed.
Playwright MCP Use Cases
- Automating data extraction from dynamic, single-page applications (SPAs).
- Performing automated UI testing and verification across different browser engines.
- Executing complex multi-step form submissions and user registration flows.
- Generating visual snapshots and accessibility trees for AI-driven page analysis.
- Capturing screenshots and videos of web interactions for audit trails and debugging.
How Playwright MCP Works
- The Openclaw Skills runner initializes the Playwright MCP server using the stdio transport layer.
- The AI agent selects the appropriate tool, such as browser_navigate, to target a specific web resource.
- Playwright launches a browser instance (headless by default) and executes the requested interaction command.
- For data extraction, the browser_evaluate tool runs custom JavaScript directly within the page context to retrieve specific nodes or values.
- Results, including text content, page snapshots, or execution logs, are returned to the agent for further processing.
Playwright MCP Setup
To integrate this into your environment, follow these installation steps:
# Install the Playwright MCP package
npm install -g @playwright/mcp
# Install the necessary browser binaries
npx playwright install chromium
# Run the MCP server to begin automation
npx @playwright/mcp
Playwright MCP Data Schema & Taxonomy
The skill utilizes structured inputs to control browser behavior. Below is a summary of the data handled by key tools within the Openclaw Skills ecosystem:
| Tool | Input Parameters | Output Type |
|---|---|---|
| browser_navigate | url (string) | Page Load Status |
| browser_click | selector (string) | Success/Failure |
| browser_type | selector (string), text (string) | Input Confirmation |
| browser_evaluate | script (string) | JSON/Primitive |
| browser_snapshot | none | DOM Structure JSON |
Playwright MCP Advanced Features
- Multi-Browser Support: Choose between chromium, firefox, or webkit engines for cross-browser compatibility.
- Domain Whitelisting: Use the --allowed-hosts flag to restrict the agent's navigation to specific trusted domains.
- Media Recording: Capability to save videos and traces of the browser session for advanced troubleshooting.
- Network Control: Configurable options to ignore HTTPS errors, block specific origins, or set custom user agents.
- Headless Toggle: Option to run the browser in headed mode for visual observation during development.
SKILL.md
Loading
Related Openclaw Skills
n8n Workflow Automation
A specialized skill for designing production-grade n8n workflows featuring built-in error handling, idempotency, and comprehensive audit logging.
Elite Longterm Memory
A robust multi-layer memory architecture for AI agents that ensures persistent context across sessions using vector search, git-notes, and a Write-Ahead Log protocol.
Backtest Expert
A professional framework for the rigorous validation and stress-testing of systematic trading strategies to ensure real-world robustness.
UI/UX Design
A comprehensive toolkit for applying modern design principles, accessibility standards, and component-driven workflows to digital products.
Playwright Scraper Skill
A high-performance web scraping tool powered by Playwright, specifically engineered to bypass Cloudflare and sophisticated anti-bot protections.
Polymarket
A comprehensive CLI toolkit for querying Polymarket prediction data, tracking market momentum, and simulated paper trading.
METADATA
Featured*
