home / mcp / mobile next mcp server
Model Context Protocol Server for Mobile Automation and Scraping (iOS, Android, Emulators, Simulators and Real Devices)
Configuration
View docs{
"mcpServers": {
"mobile-next-mobile-mcp": {
"command": "npx",
"args": [
"-y",
"@mobilenext/mobile-mcp@latest"
]
}
}
}You can run a Mobile Context Protocol (MCP) server locally to automate and develop against native iOS and Android apps on real devices, simulators, or emulators. It provides a unified, platform-agnostic interface so you can script interactions, extract data, and perform multi-step workflows without needing separate knowledge for each mobile platform.
After you add the MCP server to your client or IDE, you can instruct your AI assistant to use the available tools to automate native mobile flows, read screen data, and drive complex journeys across iOS and Android. Focus on practical tasks such as launching apps, taking screenshots, tapping at coordinates, entering text, and navigating across screens. The server uses a combination of native accessibility data and, when needed, screenshot-based analysis to determine the next action.
Prerequisites you need before installing the MCP server include Node.js, a package manager, and access to the MCP tooling you plan to use. Install Node.js and a package manager, then prepare to run the MCP server as described in the steps below.
Configuration, security, and practical usage notes help you get the most from Mobile MCP. Keep your development devices accessible, ensure your automation tools have proper permissions on each device, and follow best practices for securing your automation workflows to prevent unintended access.
Enumerates available simulators, emulators, and real devices so you can target the correct platform for automation.
Starts an application on the target device using its package name or bundle identifier.
Captures the current screen to analyze UI and extract data.
Simulates taps, long presses, and swipes at specific coordinates or based on UI element data.
Inputs text into focused fields and supports optional submission actions.
Extracts accessibility trees or view-hierarchy data to inform next actions.