The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection because it processes untrusted log data from the Unity Editor console.
Ingestion points: The console_get_logs and console_get_stats skills read arbitrary strings directly from the Unity console history.
Boundary markers: There are no explicit delimiters or instructions provided to the agent to ignore potentially malicious commands embedded within the logs.
Capability inventory: The skill includes file system write capabilities through console_export.
Sanitization: The skill does not specify any sanitization of log content before returning it to the agent, nor does it validate the savePath parameter to prevent writing outside intended directories.

unity-console