The Agent Skills Directory

[SAFE]: The skill provides guidelines for creating disciplined 'dynamic workflow harnesses'. It does not contain any executable code or instructions that bypass safety filters. Instead, it advocates for structured workflows with clear objectives and handoffs.
[SAFE]: The instructions explicitly include security best practices, such as requiring 'no-secret scans', security reviews for merge readiness, and human-in-the-loop checkpoints for workflows with safety risks.
[SAFE]: No indicators of malicious activity such as prompt injection, obfuscation, unauthorized data exfiltration, or persistence mechanisms were found in the skill content or metadata.

dynamic-workflow-mode