{"version":"2.1.293","anchor":"tool-result-turn-ending-logic-returns-endsturntouser","canonical_anchor":"tool-result-turn-ending-logic-returns-endsturntouser","heading":"The opt-in tool behaviour now says whether Claude's turn should end","tier":"use","area":"Tools","scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.293\/e\/tool-result-turn-ending-logic-returns-endsturntouser","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.293","markdown":"### The opt-in tool behaviour now says whether Claude's turn should end\n\nWith `CLAUDE_CODE_PEWTER_OWL_TOOL` set, the affected tool path now reports whether the turn ends and what to tell you, not a response ID\n\n**Unclear.** What a reader who has set `CLAUDE_CODE_PEWTER_OWL_TOOL` sees differently when the turn now ends or carries on.\n\n**What**\n\nThe behaviour you opt into with the environment variable `CLAUDE_CODE_PEWTER_OWL_TOOL` now gives a different kind of answer. It used to return a response ID, or nothing. It now returns either:\n\n- that Claude's turn should end, or\n\n- that the turn should carry on, together with a message for you.\n\nThe behaviour can also now be bypassed for particular calls. It still depends on `CLAUDE_CODE_PEWTER_OWL_TOOL` and on a remotely controlled switch, which falls back to on when the server sends nothing.\n\n**Why**\n\nThis changes how the opt-in path signals the end of Claude's turn, the point where Claude stops and hands back to you.\n\n- Flag `tengu_witty_hearth`: Not enough to say (read for one account on one subscription tier against v2.1.293; this account: no value returned, anonymous baseline: no value returned, compiled default: on) 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- Area: Tools\n- Names: `CLAUDE_CODE_PEWTER_OWL_TOOL`\n- Tier: Use it now\n- Useful: 4\/5\n- Signal: 4\/5\n- Scope: individual\n- Heads-up: no"}