Claude Code 2.1.294

如果你習慣用白話句子設定規則來管理 Claude，現在執行會更踏實。在設定 prompt 與 agent 類型的掛鉤（hook，特定事件發生時被觸發的小程式）時，你可以直接用自然語言寫下交代。以前若寫了「阻擋特定指令（Block commands that...）」等指示，系統有時會判定失誤而放行。在 Stop 與 SubagentStop 事件上交代「建置失敗請繼續（Carry on if the build is broken）」時，Claude 也常誤判而提早收工。現在這類判斷已經修好，該攔截的動作不再漏放，交代繼續時也不會輕易早退。

If you manage Claude using plain language rules, execution is now much more reliable. When configuring prompt and agent hooks (small programs triggered by specific events), you can write instructions directly in natural language. Previously, instructions like "Block commands that..." sometimes failed evaluation and allowed commands through. When telling Claude "Carry on if the build is broken" on Stop and SubagentStop events, Claude also often misjudged and stopped working early. These evaluations are now fixed: actions that should be blocked are no longer missed. Claude is less likely to stop early when instructed to continue.