{"version":"2.1.281","anchor":"context-usage-estimate-passes-after-boundary","canonical_anchor":"context-usage-estimate-passes-after-boundary","heading":"Context-usage estimate passes after_boundary","tier":"internal","area":"Elsewhere","url":"https:\/\/changelogs.core-directive.com\/v\/2.1.281\/e\/context-usage-estimate-passes-after-boundary","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.281","markdown":"### Context-usage estimate passes after_boundary\n\nA late end-of-turn context-usage estimate is now applied with an after_boundary marker\n\n**Unclear.** The finding does not say what `after_boundary` changes about how the late estimate is used.\n\n**What**\n\nAt the end of each turn Claude Code estimates how much of the conversation's space is used. When that estimate takes longer than its time limit, the late result is now applied with an `after_boundary` marker. The debug message about the slow estimate is unchanged.\n\n**Why**\n\nA slow estimate is now handled differently when it finally arrives.\n\n- Area: Elsewhere\n- Tier: Under the hood\n- Useful: 1\/5\n- Signal: 1\/5"}