{"version":"2.1.315","anchor":"code-style-line-removed-from-the-text-output-prompt-section","canonical_anchor":"code-style-line-removed-from-the-text-output-prompt-section","heading":"Line asking the model to match surrounding code style removed from one prompt option","tier":"notice","area":"System Prompt","scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.315\/e\/code-style-line-removed-from-the-text-output-prompt-section","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.315","markdown":"### Line asking the model to match surrounding code style removed from one prompt option\n\nOne option in Claude Code's instructions to the model no longer adds the line telling it to write code that matches the surrounding code\n\n**Unclear.** It is not clear whether the line was dropped outright or moved elsewhere in the prompt, so whether the model still gets this instruction is unsettled.\n\n**What**\n\nThe system prompt is the set of instructions Claude Code gives the model before your conversation starts. One part of it, about how the model should write its output, has several options.\n\nOne option, called \"short\", used to add a single line: \"Write code that reads like the surrounding code: match its comment density, naming, and idiom.\" That option is now called \"none\" and adds nothing.\n\nThe same line shows up as a change in three prompts:\n\n- the main system prompt\n\n- the system prompt used by the Agent SDK\n\n- the prompt used when compacting, which means shortening a long conversation by summarising it\n\n**Why**\n\nThis changes what the model is told about matching the style of the code around it. If edits start to follow the existing comments, naming or habits of a file less closely, this is a likely place to look.\n\n- Area: System Prompt\n- Tier: You'll notice\n- Useful: 2\/5\n- Signal: 2\/5\n- Scope: individual\n- Heads-up: no"}