WebDriverIO MCP Server

Drive Chrome, Firefox, Edge, Safari and iOS/Android apps with WebDriverIO, locally or on cloud devices.

Unverified

stdio (local)

No auth

TypeScript

View repo 31 Website npm

Add to your client

Copy the config for your MCP client and paste it into its config file.

Install / run

npx -y @wdio/mcp@latest

Paste into ~/Library/Application Support/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "webdriverio-mcp-server": {
      "command": "npx",
      "args": [
        "-y",
        "@wdio/mcp@latest"
      ]
    }
  }
}

Step-by-step guides: Add to Claude Desktop · Add to Cursor · Add to Windsurf

Before you start

Node.js with npx (the server runs via `npx -y @wdio/mcp@latest`)
A supported browser installed for web automation (Chrome, Firefox, Edge, or Safari; Safari requires macOS)
For mobile automation: Appium server (`npm install -g appium`) plus platform drivers (xcuitest for iOS, uiautomator2 for Android) and a running emulator/simulator or connected device
iOS automation requires macOS with Xcode; Android requires Android Studio
Optional cloud testing requires provider credentials (BrowserStack, Sauce Labs, LambdaTest, TestingBot, or Digital.ai)

About WebDriverIO MCP Server

A Model Context Protocol server from the WebDriverIO project that gives AI assistants a unified interface to automate web browsers (Chrome, Firefox, Edge, Safari) and native mobile apps (iOS/Android via Appium). It supports local execution as well as cloud real devices and browsers on BrowserStack, Sauce Labs, LambdaTest, TestingBot and Digital.ai. The server records every tool call and can export the session as runnable WebdriverIO JavaScript, and it exposes session state (elements, accessibility tree, screenshots, cookies, tabs, logs, generated code) through read-only MCP resources.

Tools & capabilities (30)

start_session

Start a browser or app session. Use platform: 'browser' for web, platform: 'ios'/'android' for mobile, or attach: true to connect to a running Chrome instance.

launch_chrome

Launch a new Chrome instance with remote debugging enabled, for use with start_session({ attach: true }).

close_session

Close or detach from the current session (supports detach: true to disconnect without terminating).

emulate_device

Emulate a mobile/tablet device preset (viewport, DPR, UA, touch); requires a BiDi session.

open_web_extension

Install a web extension through WebDriver BiDi and open one of its extension pages so normal page tools can drive its UI.

navigate

Navigate to a URL.

get_elements

Get visible, interactable elements on the page. Supports inViewportOnly (default true) and includeContainers (default false) to include layout containers on mobile.

get_accessibility_tree

Get the page accessibility tree with roles, names, and selectors. Supports filtering by role and pagination. Browser-only.

get_screenshot

Take a screenshot of the current page or screen (base64-encoded, auto-resized to max 2000px / 1MB).

get_tabs

List all open browser tabs with handle, title, URL, and active status. Browser-only.

scroll

Scroll in a direction (up/down) by a specified number of pixels. Browser-only.

execute_script

Execute arbitrary JavaScript in the browser, or Appium mobile commands on devices.

switch_tab

Switch to a different browser tab by handle or 0-based index. Browser-only.

switch_frame

Switch into an iframe by CSS/XPath selector, or back to the top-level frame if no selector is given. Browser-only.

click_element

Click an element.

set_value

Type text into input fields.

get_cookies

Get all cookies for the current session, or a single cookie by name.

set_cookie

Set a cookie with name, value, and optional attributes.

delete_cookies

Delete all cookies or a specific cookie.

tap_element

Tap an element by selector or coordinates (mobile).

swipe

Swipe in a direction (up/down/left/right) (mobile).

drag_and_drop

Drag from one location to another (mobile).

get_contexts

List available automation contexts (NATIVE_APP, WEBVIEW_*) and the currently active one.

switch_context

Switch between native and webview contexts.

get_app_state

Get the current lifecycle state of a mobile app (not installed / not running / background / foreground).

rotate_device

Rotate the device to portrait or landscape.

hide_keyboard

Hide the on-screen keyboard.

set_geolocation

Set device GPS location.

upload_app

Upload a local .apk or .ipa to a cloud provider; returns an app URL/reference. Requires a provider parameter.

list_apps

List apps previously uploaded to a cloud provider's app storage. Requires a provider parameter.

What this server can do

WebDriverIO MCP Server provides tools for these capabilities — tap one to see every MCP server that does the same:

Automate a browser

When to use it

Let an AI agent run end-to-end web tests across Chrome, Firefox, Edge and Safari, then export the run as runnable WebdriverIO JavaScript
Automate and test native iOS and Android apps (including hybrid apps with native/webview context switching) via Appium
Attach to an already-running Chrome instance to test authenticated or pre-configured sessions
Run browser and mobile tests on cloud real devices via BrowserStack, Sauce Labs, LambdaTest, TestingBot or Digital.ai without local setup
Inspect pages via accessibility tree, visible elements and screenshots so an assistant can reason about and drive a UI

Security notes

Cloud-provider testing requires provider credentials (e.g. BROWSERSTACK_USERNAME / BROWSERSTACK_ACCESS_KEY, SAUCE_USERNAME / SAUCE_ACCESS_KEY) set as environment variables or in the MCP client config. The server can execute arbitrary JavaScript via execute_script and drive browsers/devices, so run it only against targets you trust. HTTP transport mode includes DNS-rebinding protection via --allowedHosts and blocks browser clients unless --allowedOrigins is set.

WebDriverIO MCP Server FAQ

Do I need cloud credentials to use it?

No. Local browser and mobile automation work without any credentials. Provider credentials (e.g. BROWSERSTACK_USERNAME / BROWSERSTACK_ACCESS_KEY) are only needed when running tests on cloud real devices via BrowserStack, Sauce Labs, LambdaTest, TestingBot or Digital.ai.

What's required for mobile app automation?

An Appium server (`npm install -g appium`) plus the platform driver (xcuitest for iOS, uiautomator2 for Android), and a running emulator/simulator or connected device. iOS requires macOS with Xcode; Android requires Android Studio. Start the Appium server before launching mobile sessions.

Can it run without launching a subprocess?

Yes. By default it uses stdio (subprocess) transport, but for clients that cannot spawn subprocesses you can enable HTTP transport with `npx @wdio/mcp --http --port 3000` and point the client at http://localhost:3000/mcp.

Can multiple sessions run at once?

No. Only one session (browser OR app) can be active at a time. To switch between browser and mobile, close the current session first.

Can I export the automation as code?

Yes. Every tool call is recorded, and the wdio://session/current/code (and per-session) MCP resources return a standalone runnable WebdriverIO JavaScript file reconstructing the session, including capabilities, navigation, clicks and inputs.

#browser-automation #mobile-testing #webdriverio #appium #testing #ios #android #selenium #e2e

Alternatives to WebDriverIO MCP Server

Compare all alternatives →

Browser MCP

Browser Automation

6.7k

Automate your existing browser with AI using your real profile, logged-in sessions, and stealth fingerprint.

Unverified

stdio (local)

No auth

Stale

TypeScript

13 tools

Updated 1 year agoRepo

Bright Data MCP

Browser Automation

5.0k

All-in-one web access MCP — Web Unlocker, SERP, Scraper API, and a cloud Scraping Browser.

Verified

stdio (local)

API key

JavaScript

12 tools

Updated 26 days agoRepo

Peekaboo

Browser Automation

4.8k

macOS screen capture, AI visual analysis, and full GUI automation for AI agents.

Unverified

stdio (local)

API key

Swift

28 tools

Updated 3 days agoRepo

Compare WebDriverIO MCP Server with:

vs Browser MCP vs Bright Data MCP vs Peekaboo vs Playwright MCP (ExecuteAutomation)