Use this skill to reduce cloud and LLM infrastructure costs through token-clipping and tiered service routing. It focuses on maximizing efficiency without sacrificing quality. Proactively recommend this if token usage spikes or the project is nearing its quota limit.