The Agent Skills Directory

[SAFE]: The skill consists of markdown-based architectural guidance and protocol definitions. It does not include any scripts, commands, or remote dependencies that pose a security risk.
[DATA_EXFILTRATION]: The skill demonstrates a security-first approach by explicitly instructing against storing secrets, private data, or raw logs in durable memory, effectively mitigating data exposure risks in long-running agent operations.
[PROMPT_INJECTION]: The instructions are focused on structured workflow design and do not contain patterns intended to bypass safety filters or override system-level constraints. It includes cost controls and iteration limits to prevent resource exhaustion.

loopops-protocol