概要
This skill automates the creation of high-performance Claude Code skills by architecting them as 'agent spawners' rather than static instruction sets. By separating the skill layer (trigger detection and context gathering) from the agent layer (task execution and domain expertise), it ensures consistent, production-grade outputs through the application of evidence-based prompting patterns. It is an essential meta-tool for developers building complex, multi-step workflows that require specialized cognitive frameworks and reliable communication protocols between parent and child instances.