The Agent Skills Directory

[SAFE]: The skill contains technical documentation and code snippets for Godot Engine development. No malicious patterns, obfuscation, or unauthorized access attempts were detected.
[INDIRECT_PROMPT_INJECTION]: The skill describes common patterns for loading external game assets and mods from the local filesystem (e.g., using load() and GLTFDocument). While this establishes a data ingestion surface, it is a standard and expected feature of the Godot Engine for modding support and asset management. 1. Ingestion points: references/runtime-resource-loading.md (via the load(path) function) and SKILL.md (via the append_from_file("user://mods/enemy.glb") snippet). 2. Boundary markers: None (typical for instructional code samples). 3. Capability inventory: add_child() (attaches content to the scene tree) and change_scene_to_packed() (modifies active scene state). 4. Sanitization: None.

godot-prompter-assets-pipeline