Playwright

This Playwright MCP Server enables you to automate and control RoxyBrowser’s RoxyChrome-based browsers through AI using structured accessibility data. It offers fast, deterministic browser interactions without relying on pixel-based vision models, making it ideal for AI-driven workflows that need reliable page access and manipulation.

How to use

You can connect your MCP client to the Playwright MCP server to automate web pages using structured accessibility snapshots. Choose between a local STDIO server (running in your environment) or a remote HTTP endpoint, then issue high-level automation commands through the MCP client.

How to install

Prerequisites you need before installing:

• Node.js 18 or newer

• An MCP client such as VS Code, Cursor, Windsurf, Claude Desktop, Goose, or another MCP client

Install the Playwright MCP server using the standard configuration and then start it.

Configuration and usage notes

{
  "mcpServers": {
    "playwright": {
      "command": "npx",
      "args": ["@playwright/mcp@latest"]
    }
  }
}

Standalone and HTTP connectivity

If you run the MCP server in a headed environment and want HTTP transport, you can start it with a port and connect clients via an HTTP URL.

Example HTTP client configuration to connect to a locally hosted HTTP endpoint:

{
  "mcpServers": {
    "playwright": {
      "url": "http://localhost:8931/mcp"
    }
  }
}

Alternative STDIO configuration

To run the server locally as an STDIO process, you start the MCP server with the standard command and arguments.

{
  "mcpServers": {
    "playwright": {
      "command": "npx",
      "args": ["@playwright/mcp@latest"]
    }
  }
}

Available tools

browser_click

Perform click on a web page. Requires an element description and a page reference.

browser_close

Close the current browser page or tab.

browser_connect_roxy

Connect to a RoxyBrowser instance using a CDP WebSocket endpoint.

browser_console_messages

Retrieve all console messages from the current page.

browser_drag

Drag between two elements using their references.

browser_evaluate

Evaluate a JavaScript expression on the page or a specific element.

browser_file_upload

Upload one or multiple files to an input element.

browser_handle_dialog

Handle a dialog by choosing to accept or dismiss, with optional prompt text.

browser_hover

Hover the mouse over a target element.

browser_navigate

Navigate the browser to a specified URL.

browser_navigate_back

Go back to the previous page.

browser_navigate_forward

Go forward to the next page.

browser_network_requests

List network requests captured since the page load.

browser_press_key

Press a keyboard key or input a character.

browser_resize

Resize the browser window to specified dimensions.

browser_select_option

Select one or more options in a dropdown element.

browser_snapshot

Capture an accessibility snapshot of the current page.

browser_take_screenshot

Take a screenshot of the current viewport or element.

browser_type

Type text into an editable element.

browser_wait_for

Wait for text to appear or disappear or for a time duration to pass.

browser_tab_close

Close a specific tab by index or the current tab.

browser_tab_list

List all open tabs.

browser_tab_new

Open a new tab to a specified URL or a blank page.

browser_tab_select

Select a tab by index.

browser_install

Install the browser specified in the config.

browser_mouse_click_xy

Click at a given screen position.

browser_mouse_drag_xy

Drag the mouse from a start position to an end position.

browser_mouse_move_xy

Move the mouse to a given position.

browser_pdf_save

Save the current page as a PDF.