WallBreaker2/op

What it solves

OP provides a unified interface for Windows desktop automation, consolidating fragmented tools for window management, screen capture, input simulation, and visual recognition into a single plugin. It simplifies the process of creating scripts and automation programs that need to interact with Windows applications, especially those requiring background operation or advanced visual detection.

How it works

Implemented in C++ with bindings for Python, Go, and COM, OP integrates several low-level Windows capabilities:

• Capture & Input: Supports multiple screenshot backends (GDI, DXGI, WGC, DirectX Hook, OpenGL) and simulates mouse/keyboard input with features like smooth movement and DX input locking.
• Visual Recognition: Combines traditional color/image searching and OpenCV template matching with OCR (via dot-matrix fonts or external HTTP services like Tesseract/PaddleOCR) and YOLO object detection (via external HTTP services).
• System Interaction: Handles window enumeration, process querying, and direct memory read/write operations.

Who it’s for

Developers building Windows automation scripts, bot creators, and software testers who need a high-performance, multi-language toolkit for interacting with the OS and GUI applications.

Highlights

• Diverse Capture Backends: Supports a wide range of screenshot methods including DXGI and OpenGL ES for high-performance capture.
• Multi-Language Support: Native bindings for Python and Go, plus COM and C API access.
• Hybrid OCR/Detection: Offers both lightweight local dot-matrix OCR and scalable HTTP-based OCR and YOLO detection.
• Background Automation: Supports background binding and separated display/input handles for non-intrusive automation.
• Memory Access: Includes capabilities for reading and writing process memory.