The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the dynamic context injection syntax to execute cat commands on internal project files, such as shared protocols and configuration YAMLs. These operations are limited to reading content for prompt construction and do not involve user-controlled inputs or access to sensitive system-level data.
[DATA_EXFILTRATION]: Analysis of the skill's file access patterns shows that it only targets project-specific documentation and settings. There are no instances of credential harvesting or attempts to access sensitive files outside the project scope.
[PROMPT_INJECTION]: The instructions are focused on level design logic and architectural protocols, without any attempts to override safety filters or bypass system instructions.

level-designer