A Model Context Protocol (MCP) server for mobile automation and scraping that works with iOS, Android emulators, simulators, and real devices through a platform-agnostic interface for interacting with native applications.
One-Minute Overview#
Mobile MCP is a Model Context Protocol (MCP) server that enables mobile automation through a unified interface. It eliminates the need for distinct iOS or Android knowledge, allowing you to run automation on emulators, simulators, and real devices. The server allows Agents and LLMs to interact with native iOS/Android applications through structured accessibility snapshots or coordinate-based taps based on screenshots.
Core Value: Removes technical barriers between mobile platforms, enabling AI assistants to execute mobile device automation tasks across different platforms.
Quick Start#
Installation Difficulty: Medium - Requires installing platform tools and configuring environment, but simple via npx
# Standard config
{
"mcpServers": {
"mobile-mcp": {
"command": "npx",
"args": ["-y", "@mobilenext/mobile-mcp@latest"]
}
}
}
Is this suitable for me?
- ✅ Mobile App Automation Testing: Execute UI tests and data entry scenarios across multiple devices
- ✅ LLM-Driven User Flow Automation: Implement complex user interaction sequences through AI agents
- ✅ Cross-Platform Data Extraction: Extract structured data from mobile applications
- ❌ Non-Mobile Projects: Not suitable if your application doesn't involve mobile device interaction
Core Capabilities#
1. Device Management#
- List all available devices (simulators, emulators, and real devices)
- Get screen size and orientation information
- Set device orientation
User Value: Developers can manage multiple testing environments without manually switching devices, improving testing efficiency
2. App Management#
- List installed apps on the device
- Launch, terminate, and uninstall apps
- Install apps from files (.apk, .ipa, .app, .zip)
User Value: Implement full automation of app installation, testing, and uninstallation processes, simplifying application lifecycle management
3. Screen Interaction#
- Capture and save screenshots
- List UI elements on screen with their properties and coordinates
- Perform click, double-tap, long-press, and swipe operations
User Value: Precisely control application interface interactions, enabling complex user operation sequences
4. Input & Navigation#
- Type text with optional submission
- Press device physical buttons (HOME, BACK, VOLUME_UP/DOWN, etc.)
- Open URLs in the device browser
User Value: Simulate real user actions to complete form filling, navigation, and data entry tasks
Technology Stack & Integration#
Development Language: TypeScript, JavaScript (Node.js) Key Dependencies: Node.js v22+, Xcode command line tools, Android Platform Tools Integration Method: MCP server with support for multiple IDEs and clients
Maintenance Status#
- Development Activity: Project has a clear roadmap and ongoing development plan
- Recent Updates: Project is under active development with regular feature updates
- Community Response: Acknowledges all contributors with open-source community support
Documentation & Learning Resources#
- Documentation Quality: Comprehensive - includes installation guides for multiple IDEs/clients, example prompts, architecture diagrams, and prerequisite information
- Official Documentation: Wiki page available (referenced in README)
- Example Code: Provides multiple workflow examples such as video search, class booking, weather checking
Platform Support#
- iOS: Supports simulators and real devices via native accessibility and WebDriverAgent
- Android: Supports emulators and real devices via ADB and UI Automator
- Cross-Platform: Unified API works across both iOS and Android
System Requirements#
- Xcode command line tools
- Android Platform Tools
- Node.js v22+
- MCP supported foundational models or agents, such as Claude MCP, OpenAI Agent SDK, Copilot Studio