Maggie Appleton 写设计、人类学和智能体界面时，习惯先画清楚隐喻，再谈工具。把「Vibe Code is Legacy Code」整理成可阅读的中文笔记：问题在哪、界面默认会把人带去哪、落地时该改什么。原站导航、评论回响和广告已去掉。
问题怎么被看见
A lovely little write-up by my friend Steve Krouse on how vibe code and legacy code are roughly the same thing; “code that nobody understands.”
I particularly like this graph which illustrates the relationship between vibe code and understanding:
默认界面在强化什么
This type of discussion feels helpful in a moment where the term “vibe coding” is being tossed around in vague and unhelpful ways. It rings true to me that it’s a continuous spectrum, and no professional developers are sitting at the all-vibes end of it.
As many have pointed out , not all code written with AI assistance is vibe code. Per the original definition , it’s code written in contexts where you “forget that the code even exists.” Or as the fairly fleshed-out Wikipedia article puts it: ”A key part of the definition of vibe coding is that the user accepts code without full understanding.”
可以怎么改
Like many developers, I’m constantly grappling with how much understanding I’m willing to hand over to Cursor or Claude Code. I sincerely try to keep it minimal, or at least have them walk me through the functionality line-by-line if I feel I’m out of my depth. But it’s always easier and faster to YOLO it – an impulse I have to actively keep in check.
Our AI minions are also exceptional tools for learning when you move too far towards the high-vibes-low-understanding end of the spectrum. I particularly like getting Claude to write me targeted exercises to practice new concepts when I get lost in generated functions or fail to implement something correctly sans-AI. Even though doubling-down up on engineering skills sometimes feels like learning to operate a textile loom in 1820.
落到产品里的动作
先写出默认交互会鼓励什么行为，再决定要不要用聊天框当万能入口。没有对照的界面改动，很容易只是换皮。
落地时建议先做的 5 件事
- 先写清这个工具在强化思考还是在替人思考。
- 默认交互不要只会谄媚：该追问、该给反例、该要求证据。
- 智能体界面要露出推理步骤，而不是只给一个光滑答案。
- 知识发布用可生长的笔记，而不是一次性营销长文。
- 改产品前先画隐喻：用户以为自己在做什么，系统实际在做什么。
和智能体产品怎么接
龙虾PRO做 OpenClaw 落地时，最该从这类笔记里拿走的是「别把聊天框当唯一界面」。数字员工要能追问、能验证、能把过程摊开，而不是只负责说好话。
本文侧重智能体生命周期与技能边界。若你要动手验证，优先用 Playground 做小任务压测；权限与托管再单独规划，避免「先装一堆再治理」。 相关：Playground · 创建智能体 · 学习中心
常见问题 FAQ
什么是知识与智能体怎么？
「知识与智能体怎么」可概括为：A lovely little write-up by my friend Steve Krouse on how vibe code and legacy code are roughly the same thing; “code that nobody understands.” 本文从定义、方法与实践要点展开说明。
为什么要关注知识与智能体怎么？
关注知识与智能体怎么，是因为它直接影响效率、风险与可复制性。文中指出：A lovely little write-up by my friend Steve Krouse on how vibe code and legacy code are roughly the same thing; “code that nobody understands.”
如何落地知识与智能体怎么？有哪些关键步骤？
建议按以下路径推进知识与智能体怎么：1) 先写清这个工具在强化思考还是在替人思考。；2) 默认交互不要只会谄媚：该追问、该给反例、该要求证据。；3) 智能体界面要露出推理步骤，而不是只给一个光滑答案。；4) 知识发布用可生长的笔记，而不是一次性营销长文。；5) 改产品前先画隐喻：用户以为自己在做什么，系统实际在做什么。。细节见正文对应章节。
知识与智能体怎么适合哪些人或团队？
知识与智能体怎么更适合：产品/技术负责人、运营与增长团队、需要落地智能体或自动化的中小团队、关注「AI智能系统」方向的读者。若你只需要单次聊天式问答，可先读概念；若要上生产，请重点看步骤、权限与风控相关段落。
关于「问题怎么被看见」，本文给出了什么结论？
在「问题怎么被看见」部分，要点是：s graph which illustrates the relationship between vibe code and understanding: 默认界面在强化什么 This type of discussion feels helpful in a moment where the term “vibe coding” is being tossed around in vague and unhelpful ways.
关于「默认界面在强化什么」，本文给出了什么结论？
在「默认界面在强化什么」部分，要点是：nd no professional developers are sitting at the all-vibes end of it. As many have pointed out , not all code written with AI assistance is vibe code. Per the original definition , it’s code written in contexts where you