Use when adding monitoring, metrics, logging, or tracing to Spring AI or LLM integration code. Covers token tracking, latency measurement, cost estimation, and prompt/response logging. Use when user mentions AI monitoring, token costs, or LLM observability.
