yyds-auto
Control Android devices via MCP — tap, swipe, OCR, screenshot, UI automation, shell, file management, and AI agent orchestration for Android RPA.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/chenanzong/yyds-autoWhat This Skill Does
Yyds.Auto is a high-performance Android Robotic Process Automation (RPA) tool integrated directly into the OpenClaw ecosystem via the Model Context Protocol (MCP). It bridges the gap between LLM reasoning capabilities and physical or virtual Android device interactions. By acting as a middleware, it translates natural language instructions into precise Android actions, including gesture control (taps, swipes), OCR-based visual recognition, UI hierarchy manipulation, and file-system management. It supports both USB-connected hardware and network-connected virtual instances, making it an essential tool for cross-platform automation and on-device testing workflows.
Installation
To use Yyds.Auto with OpenClaw, ensure you have Node.js 18+ installed on your host machine. The skill is installed via the OpenClaw registry: 'clawhub install openclaw/skills/skills/chenanzong/yyds-auto'. Once the package is installed, update your MCP configuration file to include the yyds-auto server, defining the environment variables for your device host (default 127.0.0.1) and port (61140). If using USB, ensure ADB is properly authorized on your mobile device. For remote devices, verify that your local machine can reach the device IP address on the specified port.
Use Cases
- Automated Testing: Perform regression testing on Android applications by scripting complex user flows that require visual verification through OCR and UI hierarchy inspection.
- App Lifecycle Management: Automate the installation, launching, and logging of specific applications to gather data or perform routine maintenance.
- Content Aggregation: Use OCR to extract text from non-selectable UI elements within apps and pipe that data into your AI agent for analysis.
- Task Orchestration: Use the built-in AI agent capabilities to handle complex, multi-step workflows like signing into an app, navigating to a specific menu, and interacting with specific UI elements.
Example Prompts
- "Open the Instagram app, search for the 'OpenClaw' profile, and take a screenshot of the main bio page."
- "Check the current battery percentage and device network status, then list all installed third-party apps."
- "Scroll through the current feed until you find an image with text containing 'Promotion', tap on it, and extract the promotional code from the screen."
Tips & Limitations
- Permissions: Ensure 'Developer Options' and 'USB Debugging' are enabled on your Android device to allow the engine sufficient privileges.
- Performance: OCR and screenshot operations can be bandwidth-intensive; perform these actions sparingly in loops to maintain system responsiveness.
- Stability: While the engine supports root-level shell commands, prioritize using the provided MCP UI tools for interaction to minimize the risk of application crashes or state misalignment.
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-chenanzong-yyds-auto": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Flags: network-access, file-write, file-read, code-execution