learning-rate-scheduler

Quick Review

This skill has a clear, well-organized structure but critically lacks substantive content. The description is too generic and doesn't specify what learning rate scheduling capabilities are offered (e.g., step decay, exponential, cosine annealing, warmup strategies). There is virtually no task knowledge—no implementation patterns, code examples, or specific scheduler types covered. A CLI agent would struggle to invoke this skill effectively beyond recognizing the trigger words, as it provides no actionable guidance on which schedulers to use, how to configure them, or framework-specific implementation details. While learning rate scheduling is moderately novel (requiring knowledge of optimization theory and framework APIs), the skill as documented doesn't demonstrate this complexity or provide value beyond what a general LLM could offer.