The Agent Skills Directory

[SAFE]: The skill is purely instructional and contains no executable code, scripts, or external dependencies. It defines a persona for adversarial design thinking and does not interact with external networks or sensitive files.
[PROMPT_INJECTION]: No prompt injection patterns were detected. The instructions focus on logical probing and do not attempt to override system safety guidelines or extract system prompts.
[DATA_EXFILTRATION]: No data exfiltration risks were identified. While the skill encourages the agent to read the codebase to answer questions, it does not include commands to transmit data to external domains.

grill