From 0337346f03a3de47118a6cdd8126d08fffab0de2 Mon Sep 17 00:00:00 2001 From: tukuaiai Date: Mon, 15 Dec 2025 07:56:00 +0800 Subject: [PATCH] =?UTF-8?q?docs:=20=E6=89=A9=E5=B1=95'=E5=85=83=E6=96=B9?= =?UTF-8?q?=E6=B3=95'=E7=9A=84=E6=8F=8F=E8=BF=B0?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- README.md | 2 ++ 1 file changed, 2 insertions(+) diff --git a/README.md b/README.md index ec4e62a..f837669 100644 --- a/README.md +++ b/README.md @@ -58,6 +58,8 @@ PS:下面的经验并非通用,具体实践中要分场景,辩证的看 再用优化 提示词/技能 的 提示词/技能 去优化 生成提示词/技能 的 提示词/技能, +用最终的生成 提示词/技能 的 提示词/技能 生成一切可被生成提示词/技能 的 提示词/技能 所生成的提示词/技能, + 在递归中逼近理预期状态 ## 🧭 道