{"version":"2.1.64","anchor":"simplified-output-brevity-system","canonical_anchor":"simplified-output-brevity-system","heading":"Simplified Output Brevity System","tier":null,"area":null,"scope":null,"heads_up":null,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.64\/e\/simplified-output-brevity-system","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.64","markdown":"### Simplified Output Brevity System\n\nThe previous three-tier brevity system (`tengu_swann_brevity` with `strict`\/`focused`\/`polished` levels) has been replaced by a single `tengu_sotto_voce` feature flag that, when enabled, injects a streamlined output efficiency instruction. This simplifies server-side control of model verbosity.\n\n**Evidence**\n\nBrevity flag (search for `\"tengu_sotto_voce\"`)\n\n- Flag `tengu_swann_brevity`: Not enough to say (read for one account on one subscription tier against v2.1.64; this account: no value returned, anonymous baseline: no value returned, compiled default: no gate table built for this version) These values were read against a different version of Claude Code, so treat them as the nearest reading available instead of one taken on this release.\n- Flag `tengu_sotto_voce`: On for this account, and not off by default (read for one account on one subscription tier against v2.1.64; this account: on, anonymous baseline: on, compiled default: no gate table built for this version) These values were read against a different version of Claude Code, so treat them as the nearest reading available instead of one taken on this release."}