A computational framework for research and collaboration in the AE(F)C industry.

COMPAS is not a library or a plugin, but an open source, python-based framework for computational research and collaboration in architecture, engineering, fabrication and construction (AEFC). It provides a core library with data structures, algorithms, utilities and visualisation tools for AEFC applications; packages for interaction with or integration in CAD software; project templates; an easy-to-use ecosystem for the development of extensions; transparent integration of 3rd party libraries; and much more...


The goal of COMPAS is to prevent researchers from repeating themselves and reinventing the wheel over and over again, to give them access to proven methods, useful libraries, algorithms and data, and to provide a framework for creating verifiable, reproducable, and transferable results, in a well-documented and transparent format.


Computational architecture is a highly multidisciplinary field, combining research from computer science, robotics, mathematics, automation, and several other scientific areas. To deal with the different academic backgrounds, programming skills, computational experience, and best/accepted practices of its users, and to facilitate cross-topic and cross-discipline exchange and collaboration, COMPAS is designed such that is can be used cross-platform and in combination with various AE(F)C software ecosystems, or simply from the command line.

Open source

COMPAS is developed on Github by compas-dev. All compas-dev repositories are public and made availale under the MIT License. Anyone can contribute by submitting pull requests. The core library is available as a distribution package on the Python Package Index (PyPI) and via conda-forge.