{"version":"2.1.321","anchor":"static-analysis-flags-scripts-that-break-the-prompt-cache","canonical_anchor":"static-analysis-flags-scripts-that-break-the-prompt-cache","heading":"A code check now marks scripts that break the prompt cache","tier":"internal","area":"Prompt Cache","scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.321\/e\/static-analysis-flags-scripts-that-break-the-prompt-cache","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.321","markdown":"### A code check now marks scripts that break the prompt cache\n\nA new code check marks a script as breaking the prompt cache when it contains an object with a `breakPromptCache` property\n\n**Unclear.** It is not clear which tool runs this check or whether its result is ever shown to users.\n\n**What**\n\nClaude Code has a new check that looks through code for an object with a `breakPromptCache` property. When it finds one, it marks the result as breaking the prompt cache. The prompt cache is stored, already-processed conversation text that lets later requests to the model run faster and cost less.\n\n**Why**\n\nThe check points out code that would throw that saved work away.\n\n- Area: Prompt Cache\n- Tier: Under the hood\n- Useful: 2\/5\n- Signal: 3\/5\n- Scope: individual\n- Heads-up: no"}