
Android Puppeteer Mcp
Built on uiautomator2, this server turns Claude into an Android automation agent. It connects to physical devices or emulators via ADB and exposes tools for tapping coordinates, swiping, typing text, and navigating UI. The standout feature is visual element detection: take_screenshot returns annotated images with numbered overlays on interactive elements, so you can reference "tap element 5" instead of hunting for coordinates. It also handles multi-device targeting, video recording through scrcpy, and UI hierarchy parsing. Reach for this when you need Claude to drive Android UI testing, demonstrate app workflows, or automate repetitive mobile interactions without writing test scripts.
- GitHub stars
- 4
- tools
- 12
- transport
- STDIO
- last updated
- Sep 18, 2025
Tools
12 toolsPublic tool metadata for what this MCP can expose to an agent.
list_emulatorsList all available Android emulators and devices with their status and dimensions
List all available Android emulators and devices with their status and dimensions
No parameter schema in public metadata yet.
get_device_dimensionsGet the screen dimensions of a specific Android device
Get the screen dimensions of a specific Android device
No parameter schema in public metadata yet.
get_ui_elements_infoGet detailed information about all interactive UI elements on screen
Get detailed information about all interactive UI elements on screen
No parameter schema in public metadata yet.
take_screenshotCapture annotated screenshots with numbered UI element overlays
Capture annotated screenshots with numbered UI element overlays
No parameter schema in public metadata yet.
pressTap on specific coordinates with optional long press duration
Tap on specific coordinates with optional long press duration
No parameter schema in public metadata yet.
long_pressPerform long press gestures on specific coordinates
Perform long press gestures on specific coordinates
No parameter schema in public metadata yet.
press_backPress the hardware back button
Press the hardware back button
No parameter schema in public metadata yet.
swipePerform directional or custom coordinate swipes
Perform directional or custom coordinate swipes
No parameter schema in public metadata yet.
type_textType text into focused input fields with optional text clearing
Type text into focused input fields with optional text clearing
No parameter schema in public metadata yet.
scroll_elementScroll specific UI elements in any direction
Scroll specific UI elements in any direction
No parameter schema in public metadata yet.
record_videoStart screen recording with customizable quality settings
Start screen recording with customizable quality settings
No parameter schema in public metadata yet.
stop_videoStop active screen recordings and save to local storage
Stop active screen recordings and save to local storage
No parameter schema in public metadata yet.
Sponsors
Readme
mcp-name: io.github.pedro-rivas/android-puppeteer-mcp
Android Puppeteer is a lightweight, visual-first MCP (Model Context Protocol) server that enables AI agents to interact with Android devices through intelligent UI element detection and automated interactions. Built on uiautomator2, it provides comprehensive Android automation capabilities including visual element detection, touch interactions, text input, and video recording.
Features
- Visual Element Detection Automatically detects and annotates interactive UI elements with numbered overlays for precise targeting.
Details
- Repository
- github.com/pedro-rivas/android-puppeteer-mcp
- Categories
- Web & Browser Automation, Automation & Workflows, Mobile Development
- Package
android-puppeteer-mcp- Transport
- STDIO
- Registry
- MCP registry, active