{"version":"2.1.283","anchor":"policy-limits-fetch-supports-304if-none-match-and-admitting","canonical_anchor":"policy-limits-fetch-supports-304if-none-match-and-admitting","heading":"Policy limits can be fetched conditionally or taken from a host app","tier":"internal","area":"Managed Settings","url":"https:\/\/changelogs.core-directive.com\/v\/2.1.283\/e\/policy-limits-fetch-supports-304if-none-match-and-admitting","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.283","markdown":"### Policy limits can be fetched conditionally or taken from a host app\n\nPolicy limit checks now skip re-downloading unchanged data and can accept a check already fetched by a host app\n\n**Unclear.** Nothing shows which host app, if any, currently hands over a prefetched result.\n\n**What**\n\nPolicy limits are the organization-level restrictions Claude Code reads from Anthropic. Reading them has changed in two ways:\n\n- The request now asks the server whether anything changed and handles the \"not modified\" reply, instead of always downloading the full answer.\n\n- Claude Code can accept a result already fetched by a host app that launched it, but only if the organization, sign-in, address and freshness all match. Otherwise it refuses and logs the reason, such as a mismatched account or stale data.\n\n**Why**\n\nIn practice, unchanged policy data does not need to be downloaded again, and a host app can hand over a result it already has.\n\n- Area: Managed Settings\n- Tier: Under the hood\n- Useful: 2\/5\n- Signal: 3\/5"}