{"version":"2.1.120","anchor":"lean-prompt-system-gradual-rollout","canonical_anchor":"lean-prompt-system-gradual-rollout","heading":"Lean Prompt System [Gradual Rollout]","tier":null,"area":null,"scope":null,"heads_up":null,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.120\/e\/lean-prompt-system-gradual-rollout","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.120","markdown":"### Lean Prompt System [Gradual Rollout]\n\n**What**\n\nThe full lean prompt system that compresses all tool descriptions and system prompt content for reduced token usage.\n\n**Status**\n\nFeature-flagged behind `tengu_vellum_lantern` (default false), overridable with `CLAUDE_CODE_LEAN_PROMPT=1`.\n\n**Details**\n\n- Currently restricted to Opus 4.6+ and Opus 4.7 models\n- The `tengu_ochre_finch` flag controls lean memory prompts independently\n- When fully enabled, tool descriptions are 60-80% shorter while preserving essential usage information\n- Users can preview it now via the environment variable\n\n- Flag `tengu_vellum_lantern`: Off in both readings (read for one account on one subscription tier against v2.1.120; this account: off, anonymous baseline: off, 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."}