MCP Appium Server
A Model Context Protocol (MCP) server that enables mobile app automation using Appium, supporting various device interactions, element operations, and app management through a standardized protocol.
Tools
initialize-appium
Initialize an Appium driver session for mobile automation
save-ui-hierarchy
Save UI hierarchy XML to a file
close-appium
Close the current Appium driver session
appium-screenshot
Take a screenshot using Appium
tap-element
Tap on a UI element identified by a selector
send-keys
Send text input to a UI element
get-page-source
Get the XML representation of the current UI
swipe
Perform a swipe gesture on the screen
wait-for-element
Wait for an element to be visible on screen
extract-locators
Extract element locators from UI XML source
find-by-text
Generate XPath to find element by text
generate-test-script
Generate Appium test script from actions
install-app
Install an Android application APK
take-screenshot
Take a screenshot on an Android device
list-devices
List all connected Android devices
uninstall-app
Uninstall an Android application
list-installed-packages
List all installed packages on an Android device
launch-app
Launch an app on an Android device
execute-adb-command
Execute a custom ADB command
README
MCP Appium Server
A Model Context Protocol (MCP) server implementation for mobile app automation using Appium.
<a href="https://glama.ai/mcp/servers/@Rsec08/appium-mcp"> <img width="380" height="200" src="https://glama.ai/mcp/servers/@Rsec08/appium-mcp/badge" alt="Appium Server MCP server" /> </a>
Prerequisites
- Node.js (v14 or higher)
- Java Development Kit (JDK)
- Android SDK (for Android testing)
- Xcode (for iOS testing, macOS only)
- Appium Server
- Android device or emulator / iOS device or simulator
Setup
- Install dependencies:
npm install
- Install and start Appium server:
npm install -g appium
appium
- Set up Android device/emulator:
- Enable Developer Options on your Android device
- Enable USB Debugging
- Connect device via USB or start an emulator
- Verify device is connected using
adb devices
Running Tests
- Build the project:
npm run build
- Start the MCP server:
npm run dev
- In a new terminal, run the test:
npm test
Test Configuration
The example test uses the Android Settings app as a demo. To test your own app:
-
Edit
examples/appium-test.ts:- Update
deviceNameto match your device - Set
apppath to your APK file, or - Update
appPackageandappActivityfor an installed app
- Update
-
Common capabilities configuration:
const capabilities: AppiumCapabilities = {
platformName: 'Android',
deviceName: 'YOUR_DEVICE_NAME',
automationName: 'UiAutomator2',
// For installing and testing an APK:
app: './path/to/your/app.apk',
// OR for testing an installed app:
appPackage: 'your.app.package',
appActivity: '.MainActivity',
noReset: true
};
Available Actions
The MCP server supports various Appium actions:
-
Element Interactions:
- Find elements
- Tap/click
- Type text
- Scroll to element
- Long press
-
App Management:
- Launch/close app
- Reset app
- Get current package/activity
-
Device Controls:
- Screen orientation
- Keyboard handling
- Device lock/unlock
- Screenshots
- Battery info
-
Advanced Features:
- Context switching (Native/WebView)
- File operations
- Notifications
- Custom gestures
Troubleshooting
-
Device not found:
- Check
adb devicesoutput - Verify USB debugging is enabled
- Try reconnecting the device
- Check
-
App not installing:
- Verify APK path is correct
- Check device has enough storage
- Ensure app is signed for debug
-
Elements not found:
- Use Appium Inspector to verify selectors
- Check if elements are visible on screen
- Try different locator strategies
-
Connection issues:
- Verify Appium server is running
- Check port conflicts
- Ensure correct capabilities are set
Contributing
Feel free to submit issues and pull requests for additional features or bug fixes.
License
MIT
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。