cache-report

You can monitor prompt caching efficiency by asking the assistant to show the current cache report. This executes a script to parse session-specific cache metrics, displaying hit rates and token consumption statistics to help optimize performance and manage costs.

Prompt caching stores previously processed prompt tokens to reduce redundant token processing in subsequent requests. Monitoring cache hit rates helps identify whether your caching strategy is effectively lowering token consumption, which directly reduces session latency and API costs.

High latency and unexpected token costs often indicate poor prompt caching efficiency. Running a cache report parses session request data to reveal cache hit rates and token consumption patterns, helping you determine if your prompt engineering strategy needs adjustment.

Yes, you can analyze per-request token consumption by generating detailed cache analytics. The cache report breaks down token usage and costs for individual requests within the session, providing granular data to manage budgets effectively across long-running interactions.

No external dependencies are required to run a cache report. The skill operates independently using a built-in script to parse session data and display cache metrics, requiring no additional environment setup or external libraries.

The best way to optimize prompt engineering for cost management is to monitor cache hit rates and analyze token usage patterns. Reviewing detailed cache analytics allows you to identify inefficiencies and adjust your prompts to maximize cache hits and minimize redundant processing.