{"version":"2.1.261","anchor":"auto-mode-server-classifier-context-now-has-a-size-budget-wi","canonical_anchor":"auto-mode-server-classifier-context-now-has-a-size-budget-wi","heading":"Auto-mode server-classifier context now has a size budget with progressive truncation","tier":"internal","area":"Auto Mode","scope":null,"heads_up":null,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.261\/e\/auto-mode-server-classifier-context-now-has-a-size-budget-wi","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.261","markdown":"### Auto-mode server-classifier context now has a size budget with progressive truncation\n\nAuto-mode classifier context now trims progressively to stay within a byte budget.\n\nThe context sent to the server-side auto-mode classifier now has an enforced byte budget. When the context is too large, Claude Code progressively trims it: first prior-turn context, then Chrome tab data, then the git visibility cache\/remotes, then git status porcelain output. If the static, non-trimmable portion alone exceeds the budget, it logs a warning and fires the `tengu_auto_mode_context_static_over_budget` telemetry event; if the context still exceeds a harder ceiling after trimming, it is withheld entirely instead of being sent oversized.\n\n- Area: Auto Mode\n- Tier: Under the hood\n- Useful: 2\/5\n- Signal: 2\/5"}