TAuto is a state-of-the-art, high-performance automation framework designed for professional bot development and intelligent workflow execution. It combines the power of a Visual State Machine Engine with advanced Computer Vision to automate any application or game with surgical precision.
- 🚀 Evolving from ADBCapture
- ✨ Key Features
- 🏗️ Project Structure
- 🛠️ Getting Started
- 📦 Framework Components
- 📚 Documentation
- 🤝 Contributing
- 📜 License
Originally started as a simple capture tool, TAuto has been completely re-architected into a modular, event-driven engine. We've moved beyond linear scripts to a robust State Machine architecture that can handle complex scenarios, errors, and dynamic UI changes effortlessly.
Design like an Architect, Run like a Pro.
- Visual Editor: A premium WPF-based drag-and-drop designer featuring a modern Slate Dark Theme for a professional development experience.
- Hierarchical Logic: Entry actions, Exit actions, and Periodic polling within states.
- Smart Transitions: Branching based on Image presence, Text (OCR), or variable values.
- High-Precision Matching: Multi-scale template matching with adjustable confidence.
- OCR Integration: Full support for text recognition (Tesseract).
- Adaptive OCR: Otsu thresholding and auto-polarity for robust text recognition on any background.
- Fast Capture: Optimized screen mirroring and capture for low-latency feedback.
- Natural Patterns: Randomized touch points and timing to simulate human behavior.
- Cross-Platform: Native support for both Android (ADB) and Windows (SendInput/PostMessage) automation.
- High-Performance Capture: Supports Windows Graphics Capture (WGC) and Desktop Duplication API (DDA) for low-latency Windows game automation.
- Device Agnostic: Supports multiple Android versions and screen densities.
- Global Hotkeys: System-wide F5 (Start) and F8 (Stop) shortcuts for uninterrupted bot control.
- Live Debugging: Real-time execution tracing and state visualization.
- Execution Engine: Asynchronous, non-blocking runner for high-performance automation.
| Project | Description | Role |
|---|---|---|
AutoBot.App |
The Desktop App | Visual State Machine Editor, Local Debugger |
AutoBot.Server |
The API Backend | Admin Auth, License Management, Package Hosting |
AutoBot.Engine |
The Orchestrator | SignalR Hub, spawns and manages Workers |
AutoBot.WebUI |
Admin Dashboard | Vite/React UI to manage the cluster |
AutoBot.Worker |
Headless Bot Process | Executes bot logic directly on target device |
AutoBot.Platform.Windows |
Windows Driver | SendInput, PostMessage, and WGC capture |
AutoBot.Platform.Android |
Android Driver | ADB-based interaction and screen mirroring |
AutoBot.Platform.Vision |
Vision Service | OpenCV template matching and robust OCR |
TAuto.Runner |
Legacy GUI Runner | Lightweight JSON script execution engine |
- .NET 8.0 SDK
- Node.js 18+ (for WebUI)
- PostgreSQL (for Server)
- ADB (Android Debug Bridge)
- A physical Android device or Emulator with ADB debugging enabled.
# Clone the repository (including submodules)
git clone --recursive https://github.com/txqt/AutoBot.git
# Move into the folder
cd AutoBot
# Build and run the Designer App
dotnet run --project AutoBot.App/AutoBot.App.csprojExplore our detailed guides in the Docs folder:
- 🗺️ State Machine Guide: Master the art of building smart bots.
- 📐 Architecture: Deep dive into the framework internals.
- 📖 User Guide: Basic scripting and interaction reference.
- 🚀 Deployment Guide: Production setup, env vars, and HTTPS configuration.
- 🛠️ Operations Runbook: Troubleshooting and incident response.
- 🤖 AI Bot Guide: Reference for AI-assisted bot creation.
- 🔍 OCR Troubleshooting: Resolving recognition issues.
- 📝 Changelog: Stay updated with the latest releases.
Contributions are what make the open-source community such an amazing place to learn, inspire, and create. Any contributions you make are greatly appreciated.
Please see CONTRIBUTING.md for details on our code of conduct and the process for submitting pull requests.
See the list of contributors who participated in this project.
Distributed under the MIT License. See LICENSE for more information.
Developed with ❤️ by txqt.