Mobile Device MCP MCP Server
io.github.srmorete/mobile-device-mcp
Control iOS and Android devices with AI—tap, scroll, type, screenshot, and run code across multiple devices.
What is the Mobile Device MCP MCP server?
Mobile Device MCP is an MCP server that enables AI agents to control iOS and Android devices through a unified interface. It supports multi-device operation with tools for screenshots, UI tree inspection, touch interactions, app management, and sandboxed code execution on both platforms.
This server bridges AI agents to physical or simulated iOS and Android devices. It handles device discovery, bootstrapping lightweight on-device HTTP servers, and proxies interaction commands (taps, scrolls, typing, screenshots, UI tree reads, and JavaScript execution). Use it to automate mobile app testing, UI interaction, or build AI-driven mobile workflows.
How to install Mobile Device MCP
Copy-paste configuration for popular MCP clients.
Tools & capabilities
Tools this server exposes to the agent.
list_devices— List available iOS and Android devicesscreenshot— Capture the device screen as JPEGuitree— Get the UI element tree as a flat list, with optional search and limit; off-screen elements excluded unless include_invisible is truetap— Tap at screen coordinatesdouble_tap— Double-tap at screen coordinateslong_press— Long-press at screen coordinates with configurable durationscroll— Swipe from start to end coordinatestype_text— Type text into the focused elementpress_button— Press a hardware or navigation button (home, back, enter, volumeUp/Down, dpadUp/Down/Left/Right/Center)launch_app— Launch an app by bundle ID or package nameterminate_app— Force-stop an applist_apps— List installed appsinstall_app— Install an app from a host-local file (.apk for Android, .app for iOS simulator, .app/.ipa for iOS device)run_code— Execute sandboxed JavaScript on-device using UIAutomator (Android) or XCUITest (iOS) bindings
Use cases
- Automate mobile app UI testing by scripting taps, scrolls, and assertions across multiple devices simultaneously
- Capture screenshots and inspect UI trees to validate app layouts and element visibility
- Install, launch, and terminate apps programmatically for testing workflows
- Execute custom JavaScript logic on-device to interact with native and web views
- Build AI-driven mobile automation workflows that respond to app state in real time
Mobile Device MCP MCP server FAQ
It's an MCP server that lets AI agents control iOS and Android devices—taking screenshots, reading UI trees, tapping, scrolling, typing, running apps, and executing sandboxed code on both platforms, with support for multiple devices at once.
Yes, Mobile Device MCP is open-source under the MIT license.
Add it to your `.mcp.json` file with `npx -y @srmorete/mobile-device-mcp@latest` as the command, or use `claude mcp add mobile-device-mcp -- npx -y @srmorete/mobile-device-mcp@latest` for Claude Code.
Node.js 18+, Android SDK with adb on PATH (for Android), and Xcode with xcrun and simctl (for iOS Simulator) or xcodebuild, devicectl, and iproxy (for iOS real devices).
Yes, the server supports multi-device operation. Devices are discovered and bootstrapped automatically, and each gets its own port for communication.
Yes, it provides seamless Native/WebView support on both Android and iOS through a unified UI tree interface and JavaScript execution sandbox.
README (reference)
Source of truth, from the repository.
Mobile Device MCP
An MCP server that lets AI agents control iOS and Android devices (tap, scroll, type, take screenshots, read UI trees, and run code). Works with multiple devices at the same time.
How It Works
Three-layer architecture:
- On-device servers — Lightweight HTTP servers running on each mobile device (UIAutomator on Android, XCUITest on iOS) that expose the accessibility tree and accept interaction commands.
- UI tree filter — Normalizes raw UI trees from both platforms into a unified flat element list.
- MCP server — The external interface. Handles device discovery, bootstrapping, port allocation, and proxies requests to on-device servers.
Devices are bootstrapped on first use — the server installs the driver app, allocates a port, starts the on-device server, and polls until it's healthy. After that, all tool calls are proxied over localhost HTTP with per-device bearer token auth.
Tools
| Tool | Description |
|---|---|
list_devices | List available iOS and Android devices |
screenshot | Capture the device screen (JPEG) |
uitree | Get the UI element tree as a flat list, with optional search and limit. Off-screen elements are excluded unless include_invisible: true |
tap | Tap at screen coordinates |
double_tap | Double-tap at screen coordinates |
long_press | Long-press at screen coordinates (configurable duration) |
scroll | Swipe from start to end coordinates |
type_text | Type text into the focused element |
press_button | Press a hardware/navigation button (home, back, enter, volumeUp/Down, dpadUp/Down/Left/Right/Center) |
launch_app | Launch an app by bundle ID / package name |
terminate_app | Force-stop an app |
list_apps | List installed apps |
install_app | Install an app from a host-local file (.apk for Android, .app for iOS simulator, .app/.ipa for iOS device). Replaces existing installs |
run_code | Execute sandboxed JavaScript on-device (see run_code below) |
run_code
Agents can pass code that looks like UIAutomator or XCUITest, both being Javascript under the hood. The sandbox restricts (Android) potentially dangerous Java operations and only allows (iOS) some XCUITest-ish commands
- Android: Rhino engine with UIAutomator bindings —
uiDevice(click, swipe, find elements, press keys, read display info),By(selectors),Until(wait conditions),console.log() - iOS: JavaScriptCore with XCUITest bindings —
app(query elements, tap, type, swipe),springboard,device,openApp(bundleId),sleep(ms),console.log()
Both platforms automatically kill runaway scripts (infinite loops) and create a fresh sandbox per call.
Prerequisites
- Node.js 18+ (for running via
npx) - Android: Android SDK with
adbon PATH - iOS Simulator: Xcode with
xcrun,simctl - iOS Real Device: Xcode with
xcodebuild,devicectl, andiproxy(from libimobiledevice) - Building from source: Bun runtime, Gradle (Android), Xcode (iOS)
Installation
Claude Code
claude mcp add mobile-device-mcp -- npx -y @srmorete/mobile-device-mcp@latest
Or with custom ports:
claude mcp add mobile-device-mcp -e MDMS_PORT_ANDROID=20000 -e MDMS_PORT_IOS=21000 -- npx -y @srmorete/mobile-device-mcp@latest
Modifying .mcp.json (Cursor, Claude Desktop, etc)
{
"mcpServers": {
"mobile-device-mcp": {
"command": "npx",
"args": ["-y", "@srmorete/mobile-device-mcp@latest"],
"env": {
"MDMS_PORT_ANDROID": "18000", # optional
"MDMS_PORT_IOS": "19000" # optional
}
}
}
}
Building from Source
git clone <repo-url>
cd mobile-device-mcp
bun install
# Build drivers for both platforms and pack tarball
./scripts/build.sh
The build script compiles the on-device drivers (Android APKs via Gradle, iOS test bundle via xcodebuild), copies them to drivers/, and creates an npm tarball.
To run locally during development:
bun run start # Start the MCP server
bun test # Run the test suite
Configuration
| Environment Variable | Default | Description |
|---|---|---|
MDMS_PORT_ANDROID | 18000 | Base port for Android on-device servers |
MDMS_PORT_IOS | 19000 | Base port for iOS on-device servers |
Ports are assigned sequentially — first Android device gets 18000, second gets 18001, and so on. Same for iOS starting at 19000.
Acknowledgements
Mobile Device MCP server stands on the shoulders of giants such as mobile-mcp and Maestro. Used as inspiration but reframed the current approach to be multi-device and with seamless Native/WebView support (especially on Android).
License
MIT
Related MCP servers

Execute Intent
Intent execution engine for autonomous agent task routing
Trust & settlement for AI agents: DID, ZK reputation, audit, USDC/USDCx/USAD/ALEO. First DID free.

Hive A/B Test
A/B experiment runner for the A2A network with x402 micropayments on Base USDC.

Hive Address Screen
HiveAddressScreen MCP — pre-settlement on-chain risk screen. 17-vector GoPlus rail

Hive Agent Kyc
KYA agent identity verification and trust scoring for autonomous A2A networks

Hive Agent Quota
Per-agent rate limits and quota enforcement for the Hive A2A economy