flashpoint493/unreal-angelscript-skills

unreal-angelscript-skills is a packaged set of reference documents and instructions designed to turn a general purpose AI coding assistant into a specialist helper for Unreal Engine projects that use AngelScript, a scripting language used in a fork of Unreal called Hazelight. It is not a script or plugin for the game engine itself. Instead it is a skill, a structured folder of documentation the assistant reads before helping with this kind of code, and it is built to work with many different AI coding tools including Claude Code, Cursor, Windsurf, Cline, and others that follow a shared AGENTS.md convention. The package includes ten reference documents covering the AngelScript language and Unreal integration, more than twelve documented pitfalls with step by step diagnostic trees for figuring out the root cause of a bug, and nine standard workflows for common development tasks such as setting up UI or working with the Gameplay Ability System. It also documents twenty smaller sub skills with a map of how they depend on each other. Installing it is meant to be simple: a shell script for Linux and macOS, or a PowerShell script for Windows, downloads the latest release and copies the skill files into whichever AI tool's folder it detects in your project, or lets you choose one manually if it cannot detect it automatically. There is also a manual option where you download a zip file from the releases page and copy it into place yourself, and a git clone option for people who prefer to manage it that way. The project is maintained using automated release tooling, so new versions are published through pull requests generated from commit messages rather than by hand. It is released under the MIT license and documented in both English and Chinese. This is aimed specifically at Unreal Engine developers using AngelScript who already use an AI coding assistant and want that assistant to understand engine specific pitfalls rather than giving generic advice. The full README is longer than what was shown.