saranshbamania/mobile-device-mcp

Overview

MCP server for controlling mobile devices programmatically. It provides 26 tools for screenshots, UI inspection, touch interaction, and visual analysis, integrating with Anthropic Claude and Google Gemini.

Best for

Best for
Developers building agent-driven mobile testing or automation workflows

Use cases

• Automate mobile app testing via screenshots and UI inspection
• Perform touch interactions programmatically
• Integrate mobile device control into AI agent workflows

How to use

Install

npx mobile-device-mcp

Tools exposed

• list_devices
• get_device_info
• get_screen_size
• take_screenshot
• get_ui_elements
• double_tap
• long_press
• type_text
• press_key
• list_apps
• get_current_app
• get_logs
• analyze_screen
• find_element
• smart_tap
• smart_type
• suggest_actions
• visual_diff
• extract_text
• verify_screen

Tested with

Claude Code, Cursor, Windsurf, VS Code

Notes

MCP server for controlling mobile devices programmatically. It provides 26 tools for screenshots, UI inspection, touch interaction, and visual analysis, integrating with Anthropic Claude and Google Gemini.

1 stars on GitHub. Last updated 2026-03-23.

Use cases

• Automate mobile app testing via screenshots and UI inspection
• Perform touch interactions programmatically
• Integrate mobile device control into AI agent workflows

Pros

• 26 tools cover a broad range of device control actions
• Supports both Claude and Gemini for flexible AI integration
• TypeScript codebase for developer familiarity

Cons

• Low GitHub stars suggest early stage or limited adoption
• Requires MCP client setup and external AI model access
• Dependence on external APIs for visual analysis

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.