MCP Server

Active

Apache-2.0

AutoMobile MCP Server

dev.jasonpearson/auto-mobile

AI-powered mobile device automation for Android and iOS via natural language commands.

What is the AutoMobile MCP server?

AutoMobile is an MCP server that enables AI agents to control Android and iOS devices using natural language. It provides fast UX inspection, full touch injection (tap, swipe, pinch, drag), and test execution capabilities through Kotlin Accessibility Service and Swift XCTestService, allowing agents to navigate apps, reproduce bugs, and run automated tests.

AutoMobile lets AI agents interact with your mobile devices by describing what they want to do in natural language. It uses platform-native tools (adb for Android, simctl for iOS) combined with Kotlin and Swift libraries to observe app state and inject touches. This is useful for mobile engineers who want AI to help with UX exploration, bug reproduction, performance profiling, and automated UI testing.

How to install AutoMobile

Copy-paste configuration for popular MCP clients.

transport: stdio

Config generated by PluginBench — verify against the source before use.

~/Library/Application Support/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "auto-mobile": {
      "command": "bunx",
      "args": [
        "-y",
        "@kaeawc/auto-mobile"
      ]
    }
  }
}

Tools & capabilities

Tools this server exposes to the agent.

App Navigation — Navigate apps, discover screens, and map user flows
Touch Injection — Tap, swipe, pinch, drag & drop, and shake with automatic element targeting
UX Inspection — Fast observation of app state via Kotlin Accessibility Service and Swift XCTestService
Performance Profiling — Measure app startup time (cold and warm launch) and detect scroll jank
Bug Reproduction — Reproduce bugs from reports with exact steps and screenshots
UI Test Execution — Execute native Kotlin JUnit tests and Swift XCTests with device pooling and multi-device support

Use cases

Explore app UX by having an AI navigate your app and identify confusing interactions
Reproduce bugs by pasting a bug report and getting exact reproduction steps with screenshots
Create and execute UI tests by describing test scenarios in plain English
Measure app startup performance with cold and warm launch profiling
Detect scroll performance issues and dropped frames in your app

AutoMobile MCP server FAQ

What is AutoMobile?

AutoMobile is an MCP server that lets AI agents control Android and iOS devices using natural language. It combines platform tools (adb, simctl) with Kotlin and Swift libraries to enable app navigation, touch injection, UX inspection, and automated testing.

What platforms does it support?

AutoMobile supports macOS and Linux as host platforms, with the ability to control Android devices (via adb) and iOS devices (via simctl).

Is AutoMobile free?

Yes, AutoMobile is open source under the Apache 2.0 license.

How do I install it?

You can use the one-line installer (curl -fsSL https://raw.githubusercontent.com/kaeawc/auto-mobile/refs/heads/main/scripts/install.sh | bash), install via Homebrew (brew install kaeawc/tap/auto-mobile), or follow the installation guide in the documentation.

Do I need to set up Android or iOS developer tools?

Yes, you need adb for Android device control and Xcode/simctl for iOS. The interactive installer checks your host dependencies and can optionally download these tools.

What can I use it for?

You can use it to explore app UX, reproduce bugs, create UI tests, measure startup performance, and detect scroll performance issues. It's designed for mobile engineers who want AI to help with testing and quality assurance workflows.

README (reference)

Source of truth, from the repository.

AutoMobile

AutoMobile sticker

AutoMobile lets AI agents control your mobile devices using natural language. Tell an AI what you want to do, and it interacts with your Android or iOS app.

It can do all this by being an MCP server that uses standard platform tools like adb & simctl paired with additional Kotlin & Swift libraries and apps. All components are open source. The point is to provide mobile engineers with AI workflow tools to perform UX deep dives, reproduce bugs, and run automated tests.

An AI agent navigating to the Clock app, creating a new alarm

An AI agent searching YouTube and browsing results

Explore and Test

Task	What it does
Explore app UX	Navigate your app, discover screens, map user flows, identify confusing interactions
Reproduce bugs	Paste a bug report and get exact reproduction steps with screenshots
Create UI tests	Describe test scenarios in plain English, get executable test plans
Measure startup time	Profile cold and warm launch performance
Check scroll performance	Detect jank and dropped frames

How it works

🤖 Fast UX Inspection Kotlin Accessibility Service and Swift XCTestService to enable fast, accurate observations. 10x faster than the next fastest observation toolkit.
🦾 Full Touch Injection Tap, Swipe, Pinch, Drag & Drop, Shake with automatic element targeting.
♻️ Tool Feedback Observations drive the interaction loop for all tool calls.
🧪 Test Execution Kotlin JUnitRunner & Swift XCTestRunner execute tests natively handling device pooling, multi-device tests, and automatically optimizing test timing.

Get Started

You can use our interactive installer to step through all host platform requirements and configuration options. It checks host dependencies, optionally downloads Android or iOS developer tools, and configured the MCP daemon.

curl -fsSL https://raw.githubusercontent.com/kaeawc/auto-mobile/refs/heads/main/scripts/install.sh | bash

Prefer Homebrew for the CLI:

brew install kaeawc/tap/auto-mobile

or follow the installation guide.

Documentation

Full Documentation - Complete guide with examples
FAQ - Common questions answered
Design Docs - Architecture and implementation details

Contributing

Code of Conduct
Please report security vulnerabilities via GitHub
Contributing

Related MCP servers

JobsPipe

Active

Live job postings from 30+ ATS feeds and job boards, one schema. Sign in with OAuth, or use a key.

View repository →

dev.journalize/journalize

Agent-native double-entry accounting ledger with x402 micropayments

Notes

Active

A daily notebook your AI writes in — on today's page, in your entity graph, marked as its own.

View repository →

Guardian Engine

Maintained

Deterministic recipe verification engine — validates AI-generated recipes against master SOPs.

Python

MIT

View repository →

Kamy

Maintained

Document API for AI-native software: render PDFs, e-sign, PAdES-seal, and verify.

MIT

View repository →

Canonsheet

Consistent prompts for one human character across image and video models. Paid per call, x402.