Sponsored by Deepsite.site

Promptium

Created By
sycophant218 months ago
This project is a Java-based UI automation framework that integrates with Appium and utilizes LLMs (like Claude) to drive intelligent test actions on mobile devices and simulators.
Content

Promptium

LLM-Driven UI Automation with Appium and MCP

This project is a Java-based UI automation framework that integrates with Appium and utilizes LLMs (like Claude) to drive intelligent test actions on mobile devices and simulators.

Key Features

  • UI Automation with Appium
    Supports device interaction such as tap, swipe/scroll, and screenshot capture on both local simulators and real/physical devices.

  • LLM Integration for Intelligent Test Control
    Leverages large language models (LLMs) to interpret natural-language instructions and drive UI behavior.

  • Model Context Protocol (MCP) Server
    Implements an MCP server to allow communication with Claude Desktop, enabling context-aware, language-driven test flows.

  • Java-Based Architecture
    Built with Java for reliability, performance, and seamless integration with existing test ecosystems.

Technologies Used

  • Java 23+ (Minimum Required Java 17)
  • Appium
  • MCP Protocol
  • LLM (Claude Desktop integration)

Project Structure

src/
├── main/
│   └── java/
│       └── com/jadt/
│           ├── builder/       # Appium driver capabilities builder
│           ├── connector/     # Appium driver download, intiate and connection logic
│           └── domain/        # Domain Objects for images and page source
│           └── handler/       # Actions logic
│           └── util/          # Image (Screenshot) and Page Source utilities

Configration Steps (for macOS)

  1. Install Homebrew
    1. Install: /bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)"
    2. Add to System Path:
      echo 'eval "$(/opt/homebrew/bin/brew shellenv)"' >> ~/.zprofile
      eval "$(/opt/homebrew/bin/brew shellenv)" 
      
  2. Install Node: brew install node
  3. Install Appium npm install -g appium
  4. Install XCode: xcode-select --install
  5. Carthage brew install carthage (Only for Physical iOS Devices)
  6. Install XCUITest Plugin for appium appium driver install xcuitest

iOS Device Access Requirements

  1. Apple Hardware Required
  2. You must have Xcode installed from the App Store on Mac.
  3. Connect a real iOS device via USB or use a simulator.
  4. Trust the developer certs on the device (if testing on real devices).
  5. Enable Developer Mode on ios in System Settings > Privacy & Security > Developer Mode > On
  6. Enable UI Automation on ios in System Settings > Developer > Enable UI Automation (Tap/Toggle to enable)

Claude Setup

  1. Download Claude Desktop from https://claude.ai/download (Mcp not supported on claude web yet.)
  2. Setup MCP -> Settings > Developer > Edit Config
  3. Add the following code to claude_desktop_config.json under mcpServers, if you want to use the offical release.
    {
      "promptium": {
        "command": "sh",
        "args": ["-c", "curl -sL https://github.com/sycophant21/promptium/releases/latest/download/Promptium_v0.0.1.jar -o /tmp/promptium-mcp-server.jar && java -jar /tmp/promptium-mcp-server.jar"]
      }
    }
    
  4. Add the following code to claude_desktop_config.json under mcpServers, if you want to use your own compiled version.
     {
       "promptium": {
         "command": "java",
         "args": ["-jar", "Absolute-Path/to/your/compiled/jar"]
       }
     }
    
    Note: Jar can be compiled using java and maven, the instructions to create the jar are already written in the pom.xml file, just run mvn package in the project folder and you will find the jar inside "/target" folder. Use the shaded jar.
Recommend Servers
TraeBuild with Free GPT-4.1 & Claude 3.7. Fully MCP-Ready.
Zhipu Web SearchZhipu Web Search MCP Server is a search engine specifically designed for large models. It integrates four search engines, allowing users to flexibly compare and switch between them. Building upon the web crawling and ranking capabilities of traditional search engines, it enhances intent recognition capabilities, returning results more suitable for large model processing (such as webpage titles, URLs, summaries, site names, site icons, etc.). This helps AI applications achieve "dynamic knowledge acquisition" and "precise scenario adaptation" capabilities.
Tavily Mcp
Jina AI MCP ToolsA Model Context Protocol (MCP) server that integrates with Jina AI Search Foundation APIs.
Howtocook Mcp基于Anduin2017 / HowToCook (程序员在家做饭指南)的mcp server,帮你推荐菜谱、规划膳食,解决“今天吃什么“的世纪难题; Based on Anduin2017/HowToCook (Programmer's Guide to Cooking at Home), MCP Server helps you recommend recipes, plan meals, and solve the century old problem of "what to eat today"
MCP AdvisorMCP Advisor & Installation - Use the right MCP server for your needs
AiimagemultistyleA Model Context Protocol (MCP) server for image generation and manipulation using fal.ai's Stable Diffusion model.
Visual Studio Code - Open Source ("Code - OSS")Visual Studio Code
WindsurfThe new purpose-built IDE to harness magic
DeepChatYour AI Partner on Desktop
ChatWiseThe second fastest AI chatbot™
TimeA Model Context Protocol server that provides time and timezone conversion capabilities. This server enables LLMs to get current time information and perform timezone conversions using IANA timezone names, with automatic system timezone detection.
Serper MCP ServerA Serper MCP Server
EdgeOne Pages MCPAn MCP service designed for deploying HTML content to EdgeOne Pages and obtaining an accessible public URL.
Playwright McpPlaywright MCP server
BlenderBlenderMCP connects Blender to Claude AI through the Model Context Protocol (MCP), allowing Claude to directly interact with and control Blender. This integration enables prompt assisted 3D modeling, scene creation, and manipulation.
Context7Context7 MCP Server -- Up-to-date code documentation for LLMs and AI code editors
Amap Maps高德地图官方 MCP Server
MiniMax MCPOfficial MiniMax Model Context Protocol (MCP) server that enables interaction with powerful Text to Speech, image generation and video generation APIs.
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。
CursorThe AI Code Editor