{"version":"2.1.295","anchor":"cache-ttl-override-and-prompt-classifier-tweaks","canonical_anchor":"cache-ttl-override-and-prompt-classifier-tweaks","heading":"Shorter prompt caching with transcriptCacheTtl","tier":"use","area":"Prompt Cache","scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.295\/e\/cache-ttl-override-and-prompt-classifier-tweaks","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.295","markdown":"### Shorter prompt caching with `transcriptCacheTtl`\n\nSetting `transcriptCacheTtl` to 5m makes Claude Code ask for 5-minute caching when the default would be 1 hour\n\n**Unclear.** It is not stated where `transcriptCacheTtl` is set.\n\n**What**\n\nThe prompt cache is a stored copy of the conversation that the model can reuse more cheaply for a while. Two changes:\n\n- When `transcriptCacheTtl` is `5m` and the default would be 1 hour, Claude Code asks for the cache to be kept for 5 minutes instead.\n\n- In auto mode, the rule lists used to classify requests are changed before use when a certain check is on.\n\n**Why**\n\nA shorter cache time can cost less.\n\n- Area: Prompt Cache\n- Names: `transcriptCacheTtl`\n- Tier: Use it now\n- Useful: 5\/5\n- Signal: 3\/5\n- Scope: individual\n- Heads-up: no"}