#80 · Primary category: Computer Vision

MaaFramework

black-box-testing computer-vision

An automation black-box testing framework based on image recognition

Project last updated:08/28/26

GitHub Stars

4.7K

Forks

543

Contributors

113

License

LGPL-3.0

Why we included this project

Automating a desktop game or app usually means driving it through a long sequence of on-screen states, and this framework is built for exactly that. It grew out of the MaaAssistantArknights project and was rewritten as a general-purpose, low-code automation core: you describe what to look for and what to click in a pipeline protocol, and it does the rest with template matching and OCR on captured frames. The core is written in C++, but bindings for Python, Node.js, Go, Rust, and C# let teams plug it into an existing stack without rewriting everything. It targets developers building black-box test suites and bot-like automation for GUI applications, especially where there is no internal API to hook into and the screen is the only interface. Simple flows stay readable in the low-code pipeline, while custom recognizers and actions leave room for the complex cases.

Articles for this project

No articles for this project yet.

To suggest a topic or contribute an article, contact us.

Related projects in this category