The Agent Skills Directory

[SAFE]: The SKILL.md file contains purely instructional content regarding game design theory, spatial storytelling, and level architecture. It does not contain any prompt injection attempts or instructions to bypass safety filters.
[SAFE]: The script scripts/run.sh is a standard wrapper for invoking the agent. It uses a local python script (run-skill.py) located in the repository root to execute the skill. No network operations, credential harvesting, or unauthorized privilege escalation commands are present.
[SAFE]: There are no obfuscated strings, hidden URLs, or suspicious dependencies included in the skill files.

level-designer