The Agent Skills Directory

[SAFE]: No malicious obfuscation, unauthorized network activity, or credential harvesting was detected in the skill instructions.
[COMMAND_EXECUTION]: The skill utilizes the 'u' CLI (unity-cli) to interact with a Unity instance. The commands are scoped to debugging tasks such as log retrieval, state inspection, and scene management.
[PROMPT_INJECTION]: The skill was evaluated for indirect prompt injection risks as it processes external Unity console logs.
Ingestion points: u -i <instance> console get command outputs are processed by the agent.
Boundary markers: Absent; logs are read directly into the context.
Capability inventory: Includes the ability to capture screenshots, inspect scene structure, and modify components via the u CLI.
Sanitization: Not explicitly defined; the agent relies on internal safety guardrails when processing log content.
[DATA_EXPOSURE]: The skill accesses Unity console logs and editor screenshots. These operations are essential for its primary purpose of game development debugging and are performed within the local environment scope.

unity-debug