MCP Selenium Server

The MCP Selenium Server lets you automate browser actions through standardized MCP clients. You can start browser sessions, navigate pages, interact with elements, take screenshots, and more, all via a consistent MCP interface.

How to use

To use this MCP server with an MCP client, start the server and connect your client to it using the provided command or configuration snippet. The server exposes a set of browser automation tools that let you launch a browser, navigate to URLs, locate elements, perform clicks and typing, handle mouse and keyboard inputs, take screenshots, and manage file uploads.

Typical usage patterns include starting a browser session, performing a sequence of actions on a page, and then closing the session to free resources. When you configure your MCP client, reference the Selenium MCP server by its identifier and provide the command plus arguments shown in examples. You can also configure multiple MCP servers in your client if you work with more than one automation backend.

Example configuration for an MCP client pointing at the Selenium MCP server looks like this (as an MCP JSON snippet). This configuration runs the MCP server via a command that starts the Selenium MCP runtime.

Code example (configuring the Selenium MCP server in your client):

{
  "mcpServers": {
    "selenium": {
      "command": "npx",
      "args": ["-y", "@angiejones/mcp-selenium"]
    }
  }
}

Available tools

start_browser

Launches a browser session with specified browser type and options such as headless mode and additional arguments.

navigate

Directs the browser to load a specified URL.

find_element

Finds an element on the page using a locator strategy like id, css, xpath, name, tag, or class with an optional timeout.

click_element

Clicks an element located by a chosen strategy to trigger interactions.

send_keys

Types text into a focused element identified by a locator strategy, with support for timeouts.

get_element_text

Retrieves the text content of a located element.

hover

Moves the mouse to hover over a specified element.

drag_and_drop

Drags a source element and drops it onto a target element using locator strategies for both.

double_click

Performs a double-click action on a located element.

right_click

Performs a right-click (context click) on a located element.

press_key

Simulates pressing a keyboard key, such as Enter or Tab.

upload_file

Uploads a file by interacting with a file input element, supplying the file path.

take_screenshot

Captures a screenshot of the current page, with an option to save to a file path or return data.

close_session

Ends the current browser session and releases resources.