The Agent Skills Directory

[SAFE]: The skill serves as documentation for asset management in Pixi'VN projects. It provides legitimate code examples for configuring the PixiJS AssetPack and defining asset manifests.
[EXTERNAL_DOWNLOADS]: The skill describes how to reference and load media assets (images, audio, video) from external sources such as CDNs and GitHub. These references are limited to media files for game development and do not involve the execution of remote scripts or untrusted code.
[INDIRECT_PROMPT_INJECTION]: The skill establishes an attack surface for indirect injection by loading data from external URLs. However, the instructions are intended for developer configuration of trusted asset sources, and the capabilities are restricted to media loading through the PixiJS asset system.
Ingestion points: External URLs defined in src/assets/index.ts and loaded via Assets.load().
Boundary markers: The use of an alias-based manifest system decouples game logic from raw asset paths.
Capability inventory: Asset registration (Assets.init) and loading (Assets.load, Assets.loadBundle).
Sanitization: Standard PixiJS asset parsing is assumed; the skill focuses on configuration rather than runtime input sanitization.

pixi-vn-assets