{"version":"2.1.300","anchor":"prompt-cache-messages-ttl-telemetry-added-to-tengu-api-succe","canonical_anchor":"prompt-cache-messages-ttl-telemetry-added-to-tengu-api-succe","heading":"Usage data now records how long cached messages are kept","tier":"internal","area":"Prompt Caching","scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.300\/e\/prompt-cache-messages-ttl-telemetry-added-to-tengu-api-succe","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.300","markdown":"### Usage data now records how long cached messages are kept\n\nUsage data sent after each successful request now includes how long the message part of the prompt cache is kept\n\n**What**\n\nClaude Code can cache parts of a request, so repeated text does not have to be processed again. A TTL (time to live) is how long that cached copy is kept. The usage data Claude Code sends after each successful request already recorded the cache TTL and the reason for it. It now also records `prompt_cache_messages_ttl`, a separate TTL for the conversation messages in the cache.\n\n**Why**\n\nIn practice this points to messages possibly having their own cache lifetime, separate from the rest of the request.\n\n- Area: Prompt Caching\n- Tier: Under the hood\n- Useful: 2\/5\n- Signal: 3\/5\n- Scope: individual\n- Heads-up: no"}