{"version":"2.1.321","anchor":"hooks-audit-log-notes-breakpromptcache-new-hook-timing-labe","canonical_anchor":"hooks-audit-log-notes-breakpromptcache-new-hook-timing-labe","heading":"Hook log now notes hooks that break the prompt cache","tier":"internal","area":"Hooks","scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.321\/e\/hooks-audit-log-notes-breakpromptcache-new-hook-timing-labe","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.321","markdown":"### Hook log now notes hooks that break the prompt cache\n\nWhen a hook that declares it breaks the prompt cache is registered, the log line now says 'uses breakPromptCache'\n\n**What**\n\nA hook is a command or script you set up to run automatically at certain moments while Claude Code works. When Claude Code registers a hook, it writes a line to its log. If the hook declares `breaksPromptCache`, that log line now ends with 'uses breakPromptCache'.\n\nClaude Code also records one more internal timing measurement related to hooks.\n\n**Why**\n\nThe prompt cache is how Claude Code avoids resending and reprocessing the same earlier conversation on every turn, which saves time and usage. Hooks can now say that they invalidate that cache, and the log shows which ones do, so you can tell which hook is behind a cache being thrown away.\n\n- Area: Hooks\n- Names: `breaksPromptCache`\n- Tier: Under the hood\n- Useful: 1\/5\n- Signal: 2\/5\n- Scope: individual\n- Heads-up: no"}