The Agent Skills Directory

[SAFE]: The skill provides legitimate templates for Godot Engine development. The code examples for health bars, score displays, and minimaps use standard engine features and do not contain malicious commands, network requests, or unauthorized file access.\n- [INDIRECT_PROMPT_INJECTION]: The notification stack and interaction prompt logic ingest external data for display in the UI.\n
Ingestion points: The push method in references/notifications.md and the show_for method in references/interaction-prompts.md accept strings for display.\n
Boundary markers: No delimiters or safety warnings are implemented for the displayed text.\n
Capability inventory: The skill does not contain any capabilities across all scripts for command execution, file system modification, or network communication.\n
Sanitization: Input strings are converted to text for UI labels using standard casting without specialized sanitization, but the risk is negligible as the skill lacks exploitable capabilities.

hud-system