{"version":"2.1.247","anchor":"a-writing-for-the-user-prompt-block-off-unless-explicitly-tu","canonical_anchor":"a-writing-for-the-user-prompt-block-off-unless-explicitly-tu","heading":"A \"Writing for the user\" prompt block, off unless explicitly turned on","tier":"use","area":"System Prompt","scope":null,"heads_up":null,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.247\/e\/a-writing-for-the-user-prompt-block-off-unless-explicitly-tu","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.247","markdown":"### A \"Writing for the user\" prompt block, off unless explicitly turned on\n\nA writing-style prompt block can be forced on with an environment variable; otherwise it stays off.\n\n**What**\n\nA new system-prompt section named `willow_tern` tells the model how its final message should read: lead with the answer, one idea per sentence of roughly 20 words, no em-dashes or parentheticals, keep code and numbers out of prose, and no headers under about 500 words. Set `CLAUDE_CODE_WILLOW_TERN` to any truthy value to force it on. Otherwise it needs a server-side decision, and the flag it falls back to is compiled off.\n\n**Details**\n\n- Three inputs are checked in order: the environment variable, a boolean `tengu_willow_tern` key in client data (non-boolean values are ignored with a debug line), then the flag itself with fallback false.\n- Even when the flag says yes, the section only applies to models carrying the `opus_5_prompt_bundle` tag.\n- When applied, a `tengu_willow_tern_applied` event records whether client data supplied the decision.\n\n**Evidence**\n\n`- No em-dashes, no parentheticals, no arrows.`, `tengu_willow_tern_applied`\n\n- Flag `tengu_willow_tern`: Off in both readings (read for one account on one subscription tier against v2.1.247; this account: off, anonymous baseline: off, compiled default: not a boolean we can read)\n- Area: System Prompt\n- Names: `CLAUDE_CODE_WILLOW_TERN`\n- Tier: Use it now\n- Useful: 4\/5\n- Signal: 5\/5"}