An MAB-ST–Based Relational Framework for Game AI and Simulation (Runnable Python Appendix)

This material is the practical, runnable Python appendix of the MAB-ST (Markó Anna Beáta – Set Theory) framework.Its purpose is to demonstrate, through concrete, working code examples, how the MAB-ST relational world model can be applied in interactive systems, particularly in game AI and simulation contexts. This appendix is not a complete game engine nor a production-ready framework. Instead, it provides a minimal, clean, educational reference implementation that illustrates: how a “world” can be represented as a network of relations between entities, how behavior can be derived from relational patterns through a meta-level rule layer, how the system can be extended with new subsystems (e.g., reputation, factions, quests) without fragmenting the overall control logic. The package contains fully runnable Python files (no external dependencies) that build on each other to demonstrate: a minimal relational “world core” implementation, a relational NPC behavior model (the “Town Square” scenario), a classical state-machine-based contrast implementation for comparison, relational extensions (reputation, faction influence, loyalty, quest logic), and a lightweight validation and testing layer. Important reference and prerequisite:This Python appendix is built on the theoretical “core” of the MAB-ST framework. The conceptual foundations and licensing terms of MAB-ST are available under the following DOI: MAB-ST Core / Backbone / Read This First / License DOI: 10.5281/zenodo.18518738 Users are strongly advised and expected to read the documents available under the above DOI before using this appendix, as they contain: the theoretical foundation of MAB-ST, the conceptual framework and terminology, and the applicable licensing conditions. This material does not require prior knowledge of the full mathematical formalization of MAB-ST, but without reading the “core” documents, the underlying modeling paradigm and usage framework may be misunderstood.The code examples are fully runnable and suitable for: research, educational, and experimental purposes, exploring relational modeling patterns in interactive systems, and gaining practical insight into the application of the MAB-ST approach. ⚠️ License notice:This material may be used freely for research and educational purposes.Any industrial, commercial, or product-integrated use requires a separate license from the rights holder of MAB-ST.