The Agent Skills Directory

[SAFE]: The skill implements a comprehensive auditing framework that guides an agent through phases of game security review, including attack surface mapping and anti-cheat architecture evaluation.
[SAFE]: It contains explicit safety constraints for the agent, such as instructions to redact any secrets or credentials discovered during the code review process.
[SAFE]: The skill includes a self-evolution telemetry section that logs audit outcomes to a local, platform-specific directory for internal tracking and improvement, which is standard behavior for agentic tools.
[SAFE]: Analysis of the instructions and metadata revealed no evidence of prompt injection, code obfuscation, unauthorized network operations, or persistence mechanisms.

game-security