mcp-android-server-python
If you are the rightful owner of mcp-android-server-python and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.
This project provides an MCP server for automating Android devices using uiautomator2.
The MCP Android Agent is a Model Context Protocol server designed to automate Android devices through natural language commands. It integrates seamlessly with AI agents like GitHub Copilot Chat, Claude, or Open Interpreter, allowing for remote control and automation of Android devices. The server supports a wide range of functionalities, including app management, UI interactions, and device information retrieval. It is particularly useful for AI agents that require interaction with real devices, remote device control setups, automated QA tools, and UI testing. The server is built to be compatible with Python 3.13 or higher and requires the Android Debug Bridge (adb) and a connected Android device with USB debugging enabled.
Features
- Start, stop, and manage apps by package name
- Retrieve installed apps and current foreground app
- Perform UI interactions like tap, swipe, scroll, and drag
- Get device info, screen resolution, battery status, and more
- Capture screenshots or last toast messages
Tools
mcp_health
Check whether the MCP server is running normally
connect_device
Connect to Android devices and get basic information
get_installed_apps
List all installed applications and their versions and package information
get_current_app
Get information about the current front desk application
start_app
Start the application with the package name
stop_app
Stop application by package name
stop_all_apps
Stop all running applications
screen_on
Open the screen
screen_off
Close the screen
get_device_info
Obtain detailed device information (serial number, resolution, battery, etc.)
press_key
wait)
unlock_screen
Unlock the screen (open and swipe)
check_adb
Check whether ADB installs and lists connected devices
wait_for_screen_on
Wait asynchronously until the screen is on
Click
Click on the element
long_click
Long press the element
send_text
Enter text in the currently focused field (optional clear)
get_element_info
Get UI element information (text, boundaries, clickability, etc.)
swipe
Slide from one coordinate to another
wait_for_element
Wait for the element to appear on the screen
screenshot
Capture and save screenshots from the device
scroll_to
Scroll until the specified element is visible
drag
Drag the element to the specified screen position
get_toast
Get the last toast message displayed on the screen
clear_app_data
Clear user data/cache for specified apps
wait_activity
Wait until a specific event appears
