The Agent Skills Directory

[SAFE]: The skill consists of instructional content and a shell script wrapper. No malicious patterns, data exfiltration, or unauthorized command executions were found.
[COMMAND_EXECUTION]: The scripts/run.sh file is a utility script that triggers the agent via a local Python runner. It performs standard path resolution and does not include any dangerous or obfuscated commands.
[INDIRECT_PROMPT_INJECTION]: While the agent is designed to process user-provided game concepts and tasks, it lacks any powerful capabilities (like network access or file system writes) that could be exploited via indirect injection.
Ingestion points: User tasks provided via command line arguments or stdin.
Boundary markers: None defined in the skill text.
Capability inventory: None detected.
Sanitization: Not applicable as no tools are invoked.

game-designer