Back to AI game news
AI game news

Game-Design Agentic Skills Adds Bevy Support, Expanding to 181 Game-Dev AI Skills

By GameSlash

Divergent AI's October 11 v0.1.0 release adds 82 Bevy-focused AI agent skills alongside Godot guides, with ECS patterns and documented limitations.

Divergent AI expanded its Game-Design Agentic Skills (GDAS) library to the Bevy engine in version 0.1.0, dated October 11, 2026. Previously centered on Godot, the repository now describes a separate Bevy 0.20+ skill family for AI coding agents. The release concerns reusable development instructions, not a newly released game.Original release notes

Official Game-Design Agentic Skills v0.1.0 Bevy Skills release banner dated October 11, 2026
Official Bevy update banner © Divergent AI / GitHub — project artwork identifying the Bevy Skills update; the repository specifies LGPL-3.0. This is not a gameplay screenshot.

Two engines, separate skill paths

The public documentation currently lists 181 skills: 99 for Godot (one master and 98 Domain Skills) and 82 for Bevy (one master and 81 Domain Skills). These figures describe the current repository inventory, not independently measured adoption or automatically verified working game features. The developers ask users to choose an engine-specific hub such as 'bevy-master' or 'godot-master' rather than loading every skill at once.Project documentation

For Bevy projects, the new guides center on ECS ownership: deciding which Component or Resource owns state, which System updates it, and how messages or observers react. The documentation also links focused guides for input, cameras, gameplay, UI, physics and saving. A developer can install only a relevant Domain Skill when the task is narrow, or start from bevy-master for broader architecture decisions.Read bevy-master

What the expansion does not prove

The project explicitly calls the Bevy hub 'docs-first' and does not claim a complete port of the more mature Godot material. Its roadmap marks networking as blocked until November 11 and XR for later review. GameSlash has read the repository and its example skill but has not independently run Bevy builds or assessed the quality of AI-generated games using these instructions.Scope and limitations

The practical takeaway is a second engine-specific knowledge path for AI assistants used in game development, especially programmers navigating Bevy's Rust ECS model. Developers looking for broader background on the engine can also consult GameSlash's Bevy entry.

View source