章节 1：Claude 生态概览：Project、Sub-agents 与 Skills 的区别

📝 本节摘要：

在本节中，Amir 首先介绍了本期访谈的核心主题——Claude Skills（技能），并将其称为自“子智能体（Sub-agents）”以来最重要的更新。他详细梳理了 Claude 生态中三个关键概念的区别：Projects（项目）是包含自定义指令和共享知识库的协作空间，适合团队协作；Sub-agents（子智能体）主要用于代码环境，负责将复杂任务拆解分发；而 Skills（技能）则是针对特定任务的自动化工作流，用户可以像培训“初级员工”一样，为其设定严格的护栏和操作步骤，以实现高度可重复的精确输出。

[原文] [Greg Isenberg]: in this episode Amir takes us through how to use Claude skills to build digital employees We go through AB testing idea agent marketing insight agent and then we build one live together

[译文] [Greg Isenberg]: 在这一集中，Amir 将带领我们了解如何使用 Claude Skills（技能）来构建数字员工。我们会通过 A/B 测试创意智能体、营销洞察智能体，然后我们将现场一起构建一个智能体。

[原文] [Greg Isenberg]: You're going to learn about what cloud skills is why it's the biggest thing that happened since sub agents and how to actually build them yourself

[译文] [Greg Isenberg]: 你将了解到什么是 Claude Skills，为什么它是自子智能体（sub-agents）以来发生的最大事件，以及如何亲自构建它们。

[原文] [Greg Isenberg]: Amir what are we learning today

[译文] [Greg Isenberg]: Amir，我们今天要学什么？

[原文] [Amir]: Today we're going to talk about claude skills I'm going to tell you what they actually are how they're different from projects and sub agents and claude and why this matters and how you can actually apply for work

[译文] [Amir]: 今天我们要谈谈 Claude Skills。我会告诉你它们到底是什么，它们与 Projects（项目）和 Claude 中的子智能体有何不同，为什么这很重要，以及你如何将其实际应用到工作中。

[原文] [Greg Isenberg]: Okay And by the end of this episode are we are we going to be able to apply Claude skills

[译文] [Greg Isenberg]: 好的。那么在本集结束时，我们能学会应用 Claude Skills 吗？

[原文] [Amir]: 100% I'm going to show you first I want to talk about what it actually is and why it matters But I'll show you how to use existing skills in claude that they just came out with and how to create your own and how to apply for your work So whether you're in marketing in data analysis or any sort of document creation you can actually use skills to do that

[译文] [Amir]: 100% 可以。首先我要讲讲它到底是什么以及为什么它很重要，但我也会向你展示如何使用 Claude 刚刚推出的现有技能，如何创建你自己的技能，以及如何将其应用到你的工作中。所以，无论你是在从事市场营销、数据分析还是任何类型的文档创作，你实际上都可以使用 Skills 来完成。

[原文] [Greg Isenberg]: Cool Let's do it

[译文] [Greg Isenberg]: 酷，咱们开始吧。

[原文] [Amir]: Cool So first thing is I want to talk about cloud So not a lot of people are familiar with cloud projects and I want to talk about what that actually is and why it matters and how it's kind of related to skills

[译文] [Amir]: 酷。首先我想谈谈 Claude。并不是很多人熟悉 Claude Projects（项目），我想谈谈它到底是什么，为什么它很重要，以及它与 Skills 有什么关联。

[原文] [Amir]: So within Cloud AI specifically you can actually create projects and they're essentially workspaces with a set of custom instructions So this is a system prompt and it has relevant context memories and tools

[译文] [Amir]: 具体在 Claude AI 中，你实际上可以创建 Projects（项目），它们本质上是带有一组自定义指令的工作区。这就是一个系统提示词（system prompt），并且它拥有相关的上下文、记忆和工具。

[原文] [Amir]: So say for example you're part of a a broader marketing team and you want to create a project that will uh have a set of instructions to analyze marketing data for example or generate a newsletter and you want it to connect to specific tools um have relevant context and files

[译文] [Amir]: 举个例子，假设你是一个大型营销团队的一员，你想创建一个项目，其中包含一组指令来分析营销数据，或者生成一份简报，并且你希望它连接到特定的工具，拥有相关的上下文和文件。

[原文] [Amir]: So this could be a glossery of terms you use within your organization your brand guidelines depending on the task it is that you wanted to do and then also have memories generated from the chats that you have within that project So it's really great for collaboration with other team members

[译文] [Amir]: 这可能包括你们组织内部使用的术语表、品牌指南，具体取决于你想让它执行的任务，然后它还拥有在该项目中通过聊天生成的记忆。所以这对于与其他团队成员协作非常棒。

[原文] [Amir]: Now you can also use it yourself as well but really uh the ability for you to create repeatable tasks and do you know certain set of instructions with external tools and data So if I'm in a marketing team this is something that I want to look at and essentially within cloud share with my team members and create projects around it

[译文] [Amir]: 当然你也可以自己使用它，但真正的核心在于让你能够创建可重复的任务，并使用外部工具和数据执行特定的指令集。所以如果我在一个营销团队，这就是我会关注的东西，本质上是在 Claude 内部与我的团队成员共享，并围绕它创建项目。

[原文] [Amir]: All right So the only thing with projects I would say that it's important to one work with your team members to actually refine the system instructions and then always have relevant context files As your business changes the data changes you need to update it and you have to go back and constantly update these context files Um and I'll talk about why context is important in this specific session and kind of how it ranks up against skills

[译文] [Amir]: 好的。关于 Projects，我只想说重要的一点是，首先要与团队成员合作去真正完善系统指令，然后始终保持相关的上下文文件。随着业务变化、数据变化，你需要更新它，你必须回头不断更新这些上下文文件。我会详细谈谈为什么上下文在这个特定环节很重要，以及它与 Skills 相比如何。

[原文] [Amir]: Now the next part of it is sub agents With sub aents this is more relevant in cloud code specifically And I actually use uh sub aents in cloud code to spin up multiple agents And multiple agents are really great at breaking down complex multiworkflow tasks into individual tasks with specialized agents

[译文] [Amir]: 接下来的一部分是 Sub-agents（子智能体）。关于子智能体，这在 Claude Code（代码环境）中更为相关。实际上我在 Claude Code 中使用子智能体来启动多个智能体。多个智能体非常擅长将复杂的多工作流任务分解为由专门智能体处理的单个任务。

[原文] [Amir]: So what does that mean Say for example you're building a very complex feature and you want to delegate the front end to one agent and the back end to another So within the chat you can actually spin up these agents to say "Hey Claude create an agent that will work on the front end using this set of rules and then create another agent to spin up to do the back end for it."

[译文] [Amir]: 这意味着什么呢？举个例子，假设你正在构建一个非常复杂的功能，你想把前端委托给一个智能体，后端委托给另一个。所以在聊天中，你实际上可以启动这些智能体，说：“嘿 Claude，创建一个使用这套规则处理前端的智能体，然后再启动另一个智能体来为它处理后端。”

[原文] [Amir]: And what's interesting is the context is isolated to that conversation window Um and so whatever context is provided or gathered in that conversation is actually um used as an input but those agents have a set of system instructions as well

[译文] [Amir]: 有趣的是，上下文被隔离在那个对话窗口中。因此，在该对话中提供或收集的任何上下文实际上都被用作输入，但这些智能体同时也拥有一套系统指令。

[原文] [Amir]: Now where things get interesting is skills And I want to talk about kind of why this actually matters Um skills are automated workflows and tasks that you can apply globally at a project or individual level

[译文] [Amir]: 现在有趣的地方来了，那就是 Skills（技能）。我想谈谈为什么这真的很重要。Skills 是自动化的工作流和任务，你可以在项目层面或个人层面全局应用它们。

[原文] [Amir]: So whether you're an existing project you have a set of system instructions you can use skills which is an add-on or augmented skill set um within that project or individual chat and it can do a set of set of tasks create documents create PDFs analyze documents um it can actually help build MCPS for you You can use skills to create other skills or create you know visual art as well

[译文] [Amir]: 所以无论你是在现有的项目中，拥有一套系统指令，你都可以使用 Skills，这是一种附加组件或增强技能集，可以在该项目或个人聊天中使用。它可以执行一系列任务：创建文档、创建 PDF、分析文档，它甚至可以帮你构建 MCPs（模型上下文协议）。你可以用 Skills 来创建其他 Skills，或者创作视觉艺术。

[原文] [Amir]: Now when do you actually use this It's for very specialized tasks based on the constraints and guidelines and steps built by you the expert

[译文] [Amir]: 那么你究竟什么时候使用它呢？它是用于非常专门的任务，基于作为专家的你所构建的约束、准则和步骤。

[原文] [Amir]: I think uh it was Kaparthi um a couple days ago he had an excellent analogy where it's like AI is essentially your coworker or someone that like reports to you You want to train it You want to build the guidelines You know this is not verbatim but like basically what he was trying to say was yeah it's someone that you work with and you can kind of build the limit the constraints around it and guidelines on how you want it to respond to you

[译文] [Amir]: 我记得几天前 Karpathy（注：Andrej Karpathy）有一个极好的比喻，大意是 AI 本质上是你的同事或者向你汇报的人。你要训练它，你要建立准则。这不是原话，但他基本上想表达的是，是的，它是和你一起工作的人，你可以围绕它建立限制、约束，以及关于你希望它如何回应你的准则。

[原文] [Amir]: And this is kind of similar in some nature uh you can create for example let's say you are a paid media expert and you run campaigns for your clients and you want a very detailed analysis on your visit to booked appointments and what the conversion rates look like and how that attributes to the different channels you have and what's performing better than the other in terms of campaigns

[译文] [Amir]: 这在某种本质上是相似的。比如说，假设你是一名付费媒体专家，你为客户运营广告活动，你想要一份非常详细的分析，关于从访问到预约的转化情况，转化率是多少，以及这如何归因于你拥有的不同渠道，以及在广告活动方面哪个表现得更好。

[原文] [Amir]: You can create a scale that can follow a set of custom instructions but also scripts that you can build out yourself to analyze that data And I want to circle back later on why that actually is important

[译文] [Amir]: 你可以创建一个 Skill，它可以遵循一套自定义指令，同时也遵循你自己构建的脚本来分析这些数据。我想稍后再回过头来谈谈为什么这真的很重要。

[原文] [Amir]: Um but what's also interesting is that it actually only loads context when it's relevant to the task So when a project often times you have the LLM that's determining which context to retrieve and add into the conversation window and reference it Um but in this instance it's only con based on the judgment of the task whether or not it should pull relevant context and it's just relevant to exactly what you want to get done

[译文] [Amir]: 但另一个有趣的点是，它实际上只有在与任务相关时才加载上下文。在 Project（项目）中，通常是由大语言模型（LLM）决定检索哪些上下文并将其添加到对话窗口中进行引用。但在这种情况下（使用 Skills），它是基于对任务的判断来决定是否应该提取相关上下文，且仅与你想要完成的具体事项相关。

章节 2：Skills 的核心逻辑：语境管理与脚本化执行

📝 本节摘要：

在本节中，Amir 深入探讨了 Skills 背后的技术哲学。他首先引入了“语境腐烂（Context Rot）”的概念，解释了为何给予 AI 过多上下文反而会导致性能下降和幻觉增加。Skills 的核心优势在于按需加载上下文，并允许用户通过 Markdown 文件和脚本（如 Python）定义“确定性”的操作步骤。与传统 LLM 依靠概率猜测如何分析数据不同，Skills 允许用户指定精确的计算逻辑（如“将 X 列乘以 Y 列”），从而确保输出结果的精准与稳定。

[原文] [Amir]: You can create a scale that can follow a set of custom instructions but also scripts that you can build out yourself to analyze that data And I want to circle back later on why that actually is important

[译文] [Amir]: 你可以创建一个 Skill（技能），它可以遵循一套自定义指令，同时也遵循你自己构建的脚本来分析这些数据。我想稍后再回过头来谈谈为什么这真的很重要。

[原文] [Amir]: Um but what's also interesting is that it actually only loads context when it's relevant to the task So when a project often times you have the LLM that's determining which context to retrieve and add into the conversation window and reference it

[译文] [Amir]: 但另一个有趣的点是，它实际上只有在与任务相关时才加载上下文。在 Project（项目）中，通常是由大语言模型（LLM）决定检索哪些上下文并将其添加到对话窗口中进行引用。

[原文] [Amir]: Um but in this instance it's only con based on the judgment of the task whether or not it should pull relevant context and it's just relevant to exactly what you want to get done

[译文] [Amir]: 但在这种情况下（使用 Skills），它是基于对任务的判断来决定是否应该提取相关上下文，且仅与你想要完成的具体事项相关。

[原文] [Amir]: So I would say the key takeaway here is that it's repeatable instructions It's laser focused on a set of tasks pulls in context as is needed and it has the ability to run scripts or run code to perform specific functions

[译文] [Amir]: 所以我想说，这里的关键结论是，它是可重复的指令。它高度聚焦于一组任务，按需提取上下文，并且有能力运行脚本或代码来执行特定功能。

[原文] [Amir]: Why this matters because um there's a paper great paper out there called talking about context rot and how essentially um talks about how to do effective prompt engineering and how the right amount of system prompts from you know very detailed to vague and the right amount of context has a huge impact on performance

[译文] [Amir]: 为什么这很重要？因为有一篇很好的论文谈到了“语境腐烂（Context Rot）”，本质上探讨了如何进行有效的提示词工程，以及适量的系统提示词（从非常详细到模糊）和适量的上下文如何对性能产生巨大影响。

[原文] [Amir]: and as you add more context you essentially could be you know I don't want to quote on this but potentially be like degrading performance from the LLMs and likely to lead to more hallucination

[译文] [Amir]: 随着你添加更多的上下文，你本质上可能——我不想绝对引用这一点，但可能是在降低 LLM 的性能，并且更有可能导致幻觉。

[原文] [Greg Isenberg]: Exactly The more context you have the less likely you are to hallucinate Well the more Well well yes and no The more context you have you're less likely to hallucinate with the right amount of context So it's like kind of like like a like a coworker like do you want to give them all the information or just the right amount so that it doesn't bombard them to get the right task done

[译文] [Greg Isenberg]: 没错。上下文越多，产生幻觉的可能性就越小……嗯，越多……好吧，这也是也不是。拥有更多的上下文，在上下文数量适度的情况下，你确实不太可能产生幻觉。这就像对待同事一样，你是想给他们所有的信息，还是只给适量的信息，以免信息轰炸导致无法完成正确的任务？

[原文] [Amir]: Exactly So that's that's what I would really call um break it down into So I'm going to go through some examples but I want to talk about the importance of scale and why it's actually solving a real problem that I have faced myself

[译文] [Amir]: 正是如此。这就是我真正想拆解分析的。我会举一些例子，但我想先谈谈 Skill 的重要性，以及为什么它实际上解决了我自己面临的一个真正的问题。

[原文] [Amir]: So with custom skills how it works is that you essentially uh create this markdown file that explains exactly what the skill is and what it does And you can actually create reference files that it can reference back into for additional context

[译文] [Amir]: 关于自定义 Skills，它的工作原理是你本质上创建了一个 Markdown 文件，确切地解释该技能是什么以及它做什么。你还可以创建参考文件，让它可以回溯引用以获取额外的上下文。

[原文] [Amir]: So say for example you create a skill that uh applies uh XYZ's company brand guidelines to presentation and documents And this overview um essentially you know this the skill overview has a set of tasks and instructions it follows but you can also have an additional document as a reference that is an example existing brand guideline document that it can reference and it's not it's only pulling it when it needs to

[译文] [Amir]: 举个例子，假设你创建了一个 Skill，用于将 XYZ 公司的品牌指南应用于演示文稿和文档。这个概览——本质上你知道这个 Skill 概览有一组它遵循的任务和指令，但你也可以有一个额外的文档作为参考，比如现有的品牌指南文档示例，它只在需要时才提取引用它。

[原文] [Amir]: You can take it another layer and you can essentially create custom scripts as part of that scale Now um there's a great documentation by Anthropic on this and they talk about kind of how to write good scales and descriptions

[译文] [Amir]: 你还可以更进一步，本质上可以将自定义脚本作为该 Skill 的一部分。关于这一点，Anthropic 有一份很好的文档，他们谈到了如何编写好的 Skills 和描述。

[原文] [Amir]: But what's interesting is that when you are using a cloud project and you have MCPs or tools connected connectors connected the LLM is determining which tools to call based on your instructions and how to perform that task

[译文] [Amir]: 但有趣的是，当你使用 Cloud Project（云项目）并且连接了 MCPs（模型上下文协议）或工具连接器时，LLM 是基于你的指令来决定调用哪些工具以及如何执行该任务的。

[原文] [Amir]: So say for example you have a raw um like output of your meta campaign ad data or your Google ads data and you have a project in cloud that says like it's a market analyzer I want you to your instructions are to analyze this data and give me insights

[译文] [Amir]: 举个例子，假设你有一份原始的 Meta 广告活动数据或 Google Ads 数据输出，你在 Cloud 中有一个项目，设定为市场分析器，你的指令是让它分析这些数据并给我洞察。

[原文] [Amir]: The LLM is determining how to like the model is determining how to actually look at the data and perform insights and it's it's nondeterministic in a way right like it's you know it can look at it differently every single time um and you're not giving the right guidelines on how to actually take the data and analyze it

[译文] [Amir]: 此时是 LLM 在决定——模型在决定如何真正地查看数据并执行洞察，这在某种程度上是非确定性的（nondeterministic），对吧？它可能每次查看的方式都不同，而你并没有给出关于如何实际获取数据并进行分析的正确准则。

[原文] [Amir]: And I've seen this firsthand actually working with a lot of clients where like you know um a director of revops is looking at churn data new subscription data and um they'll put the file into the cloud project and it's not giving the right output of insights they're looking for

[译文] [Amir]: 我在与许多客户合作时亲眼见过这种情况，比如一位营收运营总监正在查看流失率数据、新订阅数据，他们把文件放入 Cloud 项目中，但它并没有给出他们想要寻找的正确洞察输出。

[原文] [Amir]: How this gets interesting is you can actually create scripts that are very specific So say for example um if you wanted to um have a very set of strict guidelines on how it should actually run and analyze the data then you can create that within the scale itself to say I want you to look at column X Y and Z multiply by this divided by that to the power of this to give me this insight that way it's actual functional code that's running this and it's not deterministic nondeterministic by the model itself

[译文] [Amir]: 有趣的地方在于，你实际上可以创建非常具体的脚本。举例来说，如果你想有一套非常严格的准则来规定它应该如何运行和分析数据，那么你可以在 Skill 内部创建它，说：“我要你查看 X、Y、Z 列，乘以这个，除以那个，再做个幂运算，以此给我这个洞察。”这样一来，运行它的是实际的功能代码，而不是由模型本身进行的非确定性处理。

[原文] [Amir]: Y so um so yeah that that's kind of the beauty of skills itself where uh you're able to really um bound or create the boundaries of what I should actually work towards um for um yeah for for like building out these skills

[译文] [Amir]: 是的，所以这就是 Skills 本身的美妙之处，你能够真正地界定或创建它应该努力工作的边界，用于构建这些技能。

[原文] [Amir]: So yeah you can essentially have metadata with it resources and code you can load it as needed and it kind of breaks down exactly how you should write these skills

[译文] [Amir]: 所以是的，你本质上可以拥有元数据、资源和代码，你可以按需加载它，它确切地拆解了你应该如何编写这些 Skills。

章节 3：实战演示一：Artifact Builder 与“初级员工”管理哲学

📝 本节摘要：

在本节中，Amir 开始了第一个实战演示：利用 Artifact Builder（工件生成器）技能，为营销团队创建一个 UTM 链接生成器。他解释了 Artifacts 本质上是 Claude 内部生成的实时 Web 应用程序。以此为例，Amir 和 Greg 探讨了 AI 协作的核心哲学：应将 AI 视为一名“初级员工（Junior Teammate）”。就像指导新人一样，用户不能指望 AI 一次性掌握所有背景，而是需要为其设定明确的“护栏（Guardrails）”，并采取“分批喂养（Drip-feed）”的策略逐步提供上下文，从而避免因信息过载或指令模糊导致的挫败感。

[原文] [Amir]: Now let's jump into some examples Let's do it This this the fun part How do you actually apply this

[译文] [Amir]: 现在让我们直接看一些例子。来吧，这是有趣的部分，你实际上如何应用它？

[原文] [Amir]: So the first one we're going to go through is an artifact builder So you can actually go to Claude and it's preloaded with some existing um skills So we're going to go to capabilities and essentially you'll see there's some existing skills that are preloaded

[译文] [Amir]: 我们要看的第一个是 Artifact Builder（工件生成器）。实际上你可以去 Claude，它已经预加载了一些现有的 Skills。所以我们去“Capabilities（能力）”那里，你会看到那里本质上已经预加载了一些现有的 Skills。

[原文] [Amir]: So I have created these two ones right here We'll go through them but I want to show you the ones that are already already in there So you can have an artifact builder an MCB builder and a skill creator So it's very meta You can create skills with skills

[译文] [Amir]: 我自己创建了这边的两个，我们会一一介绍，但我想先给你看那些已经在里面的。你可以看到有 Artifact Builder、MCP Builder 和 Skill Creator。这非常“元（Meta）”，你可以用 Skills 来创建 Skills。

[原文] [Amir]: So we'll go through an artifacts builder one and I'll show you an example of what that looks like So say for example you want to create a tool that is um relevant to marketers Marketers you know when they run campaigns they always have to have UTM links to do proper attribution back to their data to see okay which campaign was driving the most and when they're seeing the analytics

[译文] [Amir]: 所以我们将先通过 Artifact Builder 来看一个例子。举个例子，假设你想创建一个与营销人员相关的工具。你知道营销人员在运行广告活动时，总是必须使用 UTM 链接来进行正确的归因，以便回溯数据，查看哪个活动带来的流量最多，以及查看分析结果。

[原文] [Amir]: So here um you know I have added the artifacts builder skill please create a UTM link generator for my marketing team

[译文] [Amir]: 所以在这里，我已经添加了 Artifact Builder 这个 Skill，（输入指令）“请为我的营销团队创建一个 UTM 链接生成器”。

[原文] [Amir]: So what's happening here is that Claude is now going to reference that skill specifically that we have defined And I'll show you what that skill looks like And essentially it's reading the documentation to understand how to build components

[译文] [Amir]: 这里发生的是，Claude 现在将专门引用我们定义的那个 Skill。我会给你看那个 Skill 长什么样。本质上它正在阅读文档以了解如何构建组件。

[原文] [Amir]: Artifacts are essentially these like live apps within cloud itself that you can create very functional web apps And you can also share with your team as well So what it's doing is it's actually referencing that skill here and now creating an artifact/web app of a UTM link generator that marketing teams can use

[译文] [Amir]: Artifacts 本质上就像是 Claude 内部的实时应用程序，你可以创建功能非常完善的 Web 应用，并且你也可以与你的团队分享。所以它正在做的就是引用那个 Skill，并现场创建一个 UTM 链接生成器的 Artifact（Web 应用），供营销团队使用。

[原文] [Amir]: And you can actually just share this with the rest of your team as well or your entire team can use this as well I mean it's literally a web app It's literally a web app

[译文] [Amir]: 你实际上可以直接把它分享给团队的其他成员，或者你的整个团队都可以使用它。我是说，它就是一个字面意义上的 Web 应用。它真的是一个 Web 应用。

[原文] [Amir]: But what's interesting is we're now creating a set of specific instructions and skills We're adding a skill to this LLM now that knows has to follow this versus before you're saying "Hey code this web app." And it's kind of you're not really defining the the guard rails or the the parameters of what it should do

[译文] [Amir]: 但有趣的是，我们现在正在创建一组特定的指令和技能。我们现在给这个 LLM 添加了一个 Skill，它知道必须遵循这些指令。相比之下，以前你可能会说“嘿，把这个 Web 应用写出来”，那样你并没有真正定义它应该做什么的“护栏（guard rails）”或参数。

[原文] [Amir]: And and what happens is people get frustrated that they're not getting the right result Exactly And then they're like "Oh you know AI doesn't work for me." Exactly Exactly Exactly

[译文] [Amir]: 结果就是人们因为没有得到正确的结果而感到沮丧。

[Greg Isenberg]: 没错。

[Amir]: 然后他们就会说：“噢你知道，AI 对我没用。”

[Greg Isenberg]: 没错，没错，正是如此。

[原文] [Amir]: So this is where it gets interesting right Like you as an individual you have the opportunity to um work with Claude and skills to build exactly the skill you're looking for to do That's a repeatable task So if you as a a marketer are doing weekly tasks of reporting create a skill that can actually help you with that Just explain to it in terms of what you're looking for what you need Be very detailed if you were to assume that you're hiring someone else to do it for you

[译文] [Amir]: 所以这就是有趣的地方，对吧？作为一个个体，你有机会与 Claude 和 Skills 合作，构建完全符合你需求的 Skill 来处理那些可重复的任务。所以如果你作为一个营销人员，每周都要做报告任务，那就创建一个能真正帮到你的 Skill。你需要向它解释你在寻找什么、你需要什么，要非常详细，就像假设你正在雇佣别人为你做这件事一样。

[原文] [Greg Isenberg]: Yeah I mean it really is I mean it really is thinking about AI as a teammate Exactly Especially a junior teammate Exactly

[译文] [Greg Isenberg]: 是的，我的意思是这真的是——真的是把 AI 当作一个队友来思考。

[Amir]: 没错。

[Greg Isenberg]: 特别是一个初级队友（Junior Teammate）。

[Amir]: 没错。

[原文] [Greg Isenberg]: That you have to really give it guard rails and really give it context cuz that's how you would you know if you hired someone junior you would be like "Okay these are the tools that you're going to use." because they don't know the tools they're going to use because they're new This is the context that you need to know about our business and how we operate And then you kind of drip feed them

[译文] [Greg Isenberg]: 你真的必须给它设定护栏，真的要给它上下文。因为这就好比你雇佣了一个初级员工，你会说：“好的，这些是你将要使用的工具。”因为他们是新来的，不知道要用什么工具。“这些是你需要了解的关于我们业务以及我们如何运作的上下文。”然后你有点像是通过“滴灌（drip feed）”的方式逐步喂给他们。

[原文] [Greg Isenberg]: You don't want to overload them right Cuz then they're going to they're not going to remember everything or they might you know it might Yeah just might be overwhelming So you drip feed them the context over time

[译文] [Greg Isenberg]: 你不想让他们过载，对吧？因为那样他们会记不住所有东西，或者可能会——是的，可能会觉得不知所措。所以你随着时间推移，分批次地给他们提供上下文。

[原文] [Amir]: Exactly But what's also interesting is if you start now as these models get better and as the toolkit expands you now have this like history of like training and reference and and metadata and memories that you've created over time now then so like someone like me who uses cloud a lot I now have a lot of pre-context of like memories and experience building these projects now I know exactly how to use skills and apply it here So I think that's where it gets really interesting

[译文] [Amir]: 确实如此。但同样有趣的是，如果你现在开始，随着这些模型变得更好，随着工具包的扩展，你就拥有了这些随着时间推移创造的训练、参考、元数据和记忆的历史。就像我这样经常使用 Claude 的人，我现在有大量的预设上下文、记忆和构建这些项目的经验，所以我现在确切地知道如何使用 Skills 并将其应用在这里。我认为这就是事情变得真正有趣的地方。

[原文] [Amir]: Um I generally think scales is probably the a huge problem solver for a lot of problems I've seen firsthand working with people Like I've worked with a lot of teams right now that have actually like a lot of go to market teams that have used cloud as part of the workflow and the number one um feedback I get is the output was not on what I expected or it's incorrect

[译文] [Amir]: 我通常认为 Skills 可能是解决我亲眼所见的许多问题的巨大方案。我现在与很多团队合作，实际上有很多走向市场（Go-to-market）的团队已经将 Claude 作为工作流的一部分，而我收到的第一大反馈就是：输出不符合我的预期，或者它是不正确的。

[原文] [Amir]: There's two reasons for that One is the promp prompting is not good right the prompter the prompter it's the problem is them but the latter of it is it's also the you can prompt you know I worked on with them on right setting the right guard rails the prompts the access to tools the right retrieval of context and it still doesn't get it just right and I think this is where skills come in and solves that problem where it's just that task so you now have an artifact that's fully functional and working

[译文] [Amir]: 这有两个原因。一是提示词（prompting）写得不好，对吧？提示者——问题出在人身上。但后者是，哪怕你写好了提示词——你知道我也和他们一起工作，设置了正确的护栏、提示词、工具访问权限和正确的上下文检索，但它仍然无法完全搞定。我认为这就是 Skills 介入并解决问题的地方，因为它只专注于那个任务，所以你现在拥有了一个功能完全正常且可用的 Artifact。

章节 4：商业机会与实战演示二：A/B 测试生成器

📝 本节摘要：

在本节中，Greg 敏锐地指出了 Skills 的商业潜力——不仅仅是自用工具，更可以作为产品出售。Amir 确认了这一点，提到了 Claude 生态中正在形成的技能与插件（Plugins）库。随后，Amir 展示了第二个实战案例：A/B 测试生成器。该 Skill 利用 Firecrawl 工具抓取目标网页（如 Greg 的 humble.com），分析页面结构，并输出具体的实验方案（如调整“案例研究”板块的位置）。两人还探讨了进一步通过脚本实现自动化月度报告的可能性，展示了从单次任务向自动化服务转型的潜力。

[原文] [Greg Isenberg]: You can also sell this to other people as a as a product right

[译文] [Greg Isenberg]: 你也可以把这个当作产品卖给其他人，对吧？

[原文] [Amir]: Yeah So I think uh Claude in collaboration with someone else created this like repository of skills I I don't I I I don't want to butcher the name so I'm not going to say it but basically there is a directory of some sort with skills and plugins because they recently came up with plugins as well which is like a collection of context and pl tools and skills and prompts allin one that you can install for your cloud workflow So there is a huge opportunity for people to sell skills

[译文] [Amir]: 是的。我想 Claude 和其他人合作创建了这个类似技能库的东西。我不想把名字弄错，所以我就不说了，但基本上有一个某种形式的目录，里面有 Skills 和插件（plugins），因为他们最近也推出了插件，这就像是上下文、工具、技能和提示词的集合，全部合而为一，你可以安装到你的 Claude 工作流中。所以人们出售 Skills 有巨大的机会。

[原文] [Greg Isenberg]: Absolutely Okay Sorry What's the difference between plugins and skills

[译文] [Greg Isenberg]: 绝对是。好的，抱歉。插件（plugins）和技能（skills）有什么区别？

[原文] [Amir]: So yeah yeah plugins just came out last week which is like a plugin You you plug it in and it has ancillary features MCP access context system instructions and I think skills now I'm not don't quote me on this Cloud's been shipping Cloud's been shipping I'm having trouble keeping up You know when I was I was building this out when I was writing this out I was like I'm trying to understand what skills is And as I was actually building with it I was like it's clear to me Cuz initially my gut reaction was this is over complicating it How is this different from projects right And now I understand why

[译文] [Amir]: 是的，插件是上周刚出来的，就像一个插件。你把它插进去，它就有辅助功能、MCP 访问权限、上下文、系统指令。至于 Skills，现在不要引用我的话，Claude 更新太快了，我都快跟不上了。你知道当我构建这个、写这个的时候，我就像是在努力理解 Skills 到底是什么。而当我实际用它构建时，我就觉得我很清楚了。因为起初我的直觉反应是这把它搞得太复杂了。这和 Projects（项目）有什么不同？现在我明白了。

[原文] [Amir]: So uh yeah So we we essentially you know https idea browser.com We can go Google CPC Black Friday Cyber Monday and then generate the URL and we have a URL Boom So that's one use case Let's make it more interesting Um I am interested in finding AB testing ideas for my website

[译文] [Amir]: 所以，是的。本质上你知道 https idea browser.com。我们可以去 Google CPC 黑色星期五网络星期一，然后生成 URL，我们就有了 URL。Boom。这是通过 Artifact Builder 做的一个用例。让我们做点更有趣的。我有兴趣为我的网站寻找 A/B 测试的想法。

[原文] [Amir]: Yeah So I have a skill that essentially looks at AB test generator and what it does is that you provide a URL and it will come up with headlines or experiments for you to run for your website to increase conversions and it actually uh the skill I created the skill using the skill creator I said I'm going to give you a URL and you're going to run a framework on actually how to run good AB tests for me

[译文] [Amir]: 是的。所以我有一个 Skill，本质上是一个 A/B 测试生成器。它的作用是你提供一个 URL，它就会提出标题或实验供你在网站上运行以提高转化率。实际上，这个 Skill 是我用 Skill Creator 创建的。我告诉它，我会给你一个 URL，你要运行一个框架，关于如何真正为我运行好的 A/B 测试。

[原文] [Amir]: So we're going to test this and see what it looks like and then I'll show you an example of how to create your own skill as well So hey Claude I have just added I have added the AB test generator skill Can you run an can you provide me with AB experiment ideas for humble.com

[译文] [Amir]: 所以我们将测试一下，看看它是什么样子，然后我也会给你展示一个如何创建你自己 Skill 的例子。所以，嘿 Claude，我刚刚添加了 A/B 测试生成器 Skill。你能运行——你能为 humble.com 提供 A/B 实验想法吗？

[原文] [Amir]: And what this will do is here because I have access to uh an FCB called firecrawl it would actually use firecrawl to scrape the URL the page and the contents and then come back with a very clear framework on experiments to run

[译文] [Amir]: 它在这里会做的是，因为我有权访问一个名为 Firecrawl 的工具（原文口误为FCB），它实际上会使用 Firecrawl 来抓取 URL、页面和内容，然后反馈一个非常清晰的实验运行框架。

[原文] [Amir]: So while that's running maybe I'll just show you an example of what that actually looks like And essentially it looks something like this So it gives you an experiment pipeline impact confidence ease um IC score

[译文] [Amir]: 在它运行的同时，也许我可以直接给你看一个例子，看看它实际长什么样。本质上它看起来像这样。它给了你一个实验流程、影响、信心、简易度……IC 评分（ICE Score）。

[原文] [Amir]: And um you know it was actually a really good one I actually did it right before this call before the session today was it asked me to it told me to actually test um shifting the case study that I have above one section above So it's like hero section and then case study go to immediately social proof and I was like damn that's a good idea like why am I showing the features when I should show do the social proof

[译文] [Amir]: 你知道这实际上是一个非常好的建议。我在今天这个会议通话之前刚做过一次，它让我去测试将我有的案例研究向上移动一个板块。所以就像是 Hero（首屏）部分，然后是案例研究，直接进入社会证明。我就想，该死，这真是个好主意，为什么我要展示功能，明明我应该展示社会证明？

[原文] [Amir]: So I'm running AB test right now to see which one is likely to drink drive more conversions and signups So um that's interesting like it really breaks down exactly the control the variant the headlines you should be testing So experiment number one experiment number two

[译文] [Amir]: 所以我现在正在运行 A/B 测试，看看哪一个更可能驱动更多的转化和注册。这很有趣，它真正确切地拆解了对照组、变体、你应该测试的标题。所以这是实验一，实验二。

[原文] [Greg Isenberg]: And if I really wanted to you know just take this put it into my app and then run run through an experiment You know would be really cool if you can automate this so that you you know every month send me a report

[译文] [Greg Isenberg]: 如果我真的想——你知道，直接把这个放进我的 App 里，然后运行实验。你知道如果能自动化这个会很酷，这样你知道每个月发给我一份报告。

[原文] [Greg Isenberg]: Yeah What to change

[译文] [Greg Isenberg]: 是的，告诉我该改什么。

[原文] [Amir]: Exactly Exactly So um if you really wanted to yeah you can You can probably write a skill uh that I wonder if you can I wonder I wonder if you can already do that today where you write a skill that writes a script that automatically sends you rapport every single week or every month

[译文] [Amir]: 没错，没错。所以如果你真的想做，是的，你可以。你大概可以写一个 Skill——我想知道你是否可以——我想知道你今天是否已经可以做到，写一个 Skill，编写一个脚本，每周或每月自动给你发送报告。

[原文] [Greg Isenberg]: Yeah Why not Yeah Yeah Yeah So yeah you know I am going to plug in my app and say we do that automatically in our app So every week we we have like four sets of sub agents that go through your website and give you insights from like a copy conversion marketing like um a designer as well So every week and then we give you like an optimization score

[译文] [Greg Isenberg]: 是的，为什么不呢？是的，是的，是的。所以你知道，我要植入一下我的 App，在我们的 App 里我们是自动做这个的。每周我们有四组子智能体（sub-agents）浏览你的网站，给你提供来自文案、转化、营销以及设计师视角的洞察。所以每周我们都会给你一个优化评分。

章节 5：实战演示三：数据分析与解决幻觉问题

📝 本节摘要：

在本节中，Amir 展示了他最期待的一个应用场景：流量数据分析。他指出，直接将 CSV 文件上传到普通 Project 中让 AI 分析，往往会导致数据计算错误或“幻觉”。而通过 Skills，他可以强制 AI 运行特定的 Python 脚本来执行计算（如总支出、净利润等），并引用 metrics.md 文件来严格定义指标含义。这种“代码级”的约束确保了输出结果的准确性和确定性，让 Greg 惊叹其产出质量达到了“产品经理”级别的精确度。

[原文] [Amir]: Now what I really really really want to show you is a problem I've been trying to solve for the past couple months with these companies I've been working with which is take data and give me the insights that I actually want to look at

[译文] [Amir]: 现在我真的非常、非常、非常想给你看一个我在过去几个月与这些公司合作时一直试图解决的问题，那就是提取数据并给我我真正想看的洞察。

[原文] [Amir]: It's such a repeatable task and it's so important that I think I can't confirm yet skills has probably solved that problem for me now in a way

[译文] [Amir]: 这是一个如此可重复的任务，而且如此重要，我觉得——虽然我还不能完全确认——Skills 可能在某种程度上已经为我解决了这个问题。

[原文] [Amir]: So um I uploaded a file called traffic analytics It it's just basically like a a CSV of um just a bunch of campaigns and you know revenue data and whatever And I was like I need some insights on this And that's what really matters to a lot of people in terms of just did cost go down did you know CBC go up down What does the trial conversion look like XYZ

[译文] [Amir]: 所以，我上传了一个名为“流量分析”的文件。它基本上就是一个 CSV 文件，包含一堆广告活动、收入数据之类的。我就想，我需要关于这个的一些洞察。这对很多人来说真的很重要，比如成本下降了吗？点击成本（CPC）上升还是下降了？试用转化率怎么样？诸如此类。

[原文] [Amir]: So I um I provided a file and it referenced the scale and has a set of scripts within that scale to then do a comprehensive analysis of the data the traffic data

[译文] [Amir]: 所以我提供了一个文件，它引用了这个 Skill，并且在这个 Skill 内部有一组脚本，用来对这些数据——流量数据——进行全面的分析。

[原文] [Amir]: So overall performance your total spend was 400k your revenue was 854K uh net profit conversions which channel did better than the other

[译文] [Amir]: 结果显示总体表现：你的总支出是 40 万，收入是 85.4 万，还有净利润、转化率、哪个渠道比另一个表现更好。

[原文] [Amir]: Um so you have a clear idea and I'll be honest like I you know you know I am going to be honest but um I would say that I I I would I would say that if I had done this through a project and I just uploaded a file with set of instructions without running scripts it would have probably hallucinated some of the data

[译文] [Amir]: 所以你有了一个清晰的概念。老实说——你知道我会实话实说的——我想说，如果我是通过 Project（项目）来做这件事，只是上传一个文件并附上一组指令，而不运行脚本，它很可能会对部分数据产生幻觉。

[原文] [Greg Isenberg]: That's what I was going to say because I when I look at this this feels this if this wasn't in in Claude and I had a product manager send me this I would be like "Yeah you know this feels like that level of fidelity." Um it just you know it just it it's it looks right It looks right It looks right

[译文] [Greg Isenberg]: 这正是我要说的。因为当我看这个的时候，感觉——如果这不是在 Claude 里，而是一个产品经理发给我这个，我会说：“是啊，你知道这感觉就是那种保真度。”它就是——你知道，它看起来是对的。看起来是对的。看起来是对的。

[原文] [Amir]: I mean I can't confirm cuz I don't know the I didn't look at the Excel but I'm just just looking at it looks right

[译文] [Amir]: 我的意思是，我不能完全确认，因为我不知道——我没看 Excel 原表，但光看这个，感觉是对的。

[原文] [Amir]: Yeah exactly So and and I'll show you what it looks like essentially within the breakdown of the skill itself So skills um what you do is you have the actual skill.md file itself So this again is a breakdown of what the skill is the scripts it should run and then yeah like generate data for 90 days generate data of 7 days generate data for 10 campaigns and then what the structure should look like

[译文] [Amir]: 是的，没错。我会给你看它在 Skill 本身的拆解中本质上是什么样子的。所以在 Skills 中，你有一个实际的 skill.md 文件。这也同样是关于该技能是什么、它应该运行什么脚本的拆解，比如生成 90 天的数据、生成 7 天的数据、生成 10 个广告活动的数据，以及结构应该长什么样。

[原文] [Amir]: So you can actually use this to define it And you know if I want to take a step further I can use cursor to then update the scale itself Um and I'll show you an example of how to create your own scale

[译文] [Amir]: 所以你实际上可以用这个来定义它。而且你知道，如果我想更进一步，我可以用 Cursor（代码编辑器）来更新这个 Skill 本身。我会给你展示一个如何创建你自己 Skill 的例子。

[原文] [Amir]: And then you can also reference files So you can see here it says see references metrics MD for detailed metric definition and typical ranges So if we want to go back into references we can see what metrics MD has which is all the you know definitions or glossery

[译文] [Amir]: 然后你还可以引用文件。所以你可以看到这里写着“参见参考资料 metrics.md 以获取详细的指标定义和典型范围”。如果我们回到参考资料中，我们可以看到 metrics.md 里有什么，也就是所有的定义或术语表。

[原文] [Amir]: So as a marketer you get to define what these are and you should be doing that so that you know when you run these scripts and skills it gives you exactly what you need instead of getting the LLM to actually do it for you

[译文] [Amir]: 所以作为一个营销人员，你可以定义这些是什么。而且你应该这样做，这样当你运行这些脚本和 Skills 时，你知道它给你的正是你需要的，而不是让 LLM 实际上去替你（猜测着）做这件事。

[原文] [Amir]: And then the scripts are made by cloud itself where it's running a Python script on you know calculating all this for you So it's accurate in some way or another

[译文] [Amir]: 然后这些脚本是由 Claude 自己生成的，它运行一个 Python 脚本来为你计算所有这些。所以在某种程度上它是准确的。

章节 6：现场构建：推文转长文 Skill 及 AI 采用率的未来

📝 本节摘要：

在访谈的最后部分，Greg 提出了一个个人需求：能否创建一个 Skill，将他每天发布的推文自动扩展为符合他写作风格的长篇 Newsletter。两人现场进行了实验，导入了 Greg 的过往推文和 Newsletter 作为风格参考（Reference）。生成的初稿质量令 Greg 大为惊叹，称其为“Banger（炸裂之作）”。

>

随后，Amir 引用了 Ramp 的一份报告，指出近期 AI 工具订阅量有所下降。他认为这并非因为 AI 无论是企业端还是个人端不再重要，而是因为大众缺乏“AI 流畅度（AI Fluency）”和提示词教育。他总结道，只要解决了教育和上下文管理的问题，AI 的生产力价值将重新推动采用率回升。

[原文] [Amir]: Yep Cool Um let's see where we are at Yeah So let's now create we we've gone through AB testing ideas We created an artifact We got some working insights I think we should now just create our own skill Do you have anything in mind Tell me if this is possible

[译文] [Amir]: 是的，酷。让我们看看我们现在的进度。是的，我们已经过了一遍 A/B 测试的想法，我们创建了一个 Artifact（工件），我们得到了一些有效的洞察。我想我们现在应该直接创建我们自己的 Skill 了。你有什么想法吗？告诉我这是否可行。

[原文] [Greg Isenberg]: So I tweet every day and I also have a newsletter and every single week I basically I use my tweets like if it rips on Twitter I'll kind of expand on it on my newsletter

[译文] [Greg Isenberg]: 我每天都发推文，我也有一个 Newsletter（时事通讯）。每周我基本上——如果不推文在 Twitter 上火了，我就会在我的 Newsletter 里对它进行扩展。

[原文] [Greg Isenberg]: Okay Um I have a specific type of style how I write on my newsletter So what would be really cool like this is some this is something I would hire for potentially like almost like a ghost writer So is it possible to have a skill that basically like looks at my tweets and turns it into long form content that I can review and and be the editor of

[译文] [Greg Isenberg]: 好的。我在 Newsletter 上的写作有一种特定的风格。所以如果能这样就太酷了——这是某种我可能会雇人来做的事情，就像雇一个代笔作家（Ghost Writer）。所以有没有可能有一个 Skill，基本上就是看我的推文，把它变成长篇内容，然后我可以审阅并担任编辑？

[原文] [Amir]: Okay let's try Let's let's let's find out Um we're doing a lot You're like "Maybe." Yeah Yeah Yeah No absolutely I think I think we can figure out we can we can try So hey Claude I just added the skill creator skill So we're using the skill creator to do that Can you make me a skill that takes an existing tweet provided by the user and turns it into long form content for LinkedIn for uh for newsletter newsletter for newsletter

[译文] [Amir]: 好的，让我们试试。让我们——让我们找出答案。如果你觉得这有点悬——不，绝对可以，我想我们可以搞定，我们可以试试。所以，嘿 Claude，我刚刚添加了 Skill Creator（技能创建者）这个 Skill。我们正在使用 Skill Creator 来做这件事。（输入指令）“你能为我制作一个 Skill 吗？它接受用户提供的现有推文，并将其转换为用于 LinkedIn……呃，用于 Newsletter 的长篇内容。”

[原文] [Amir]: Okay So I would I think what what I would do in this scenario is making sure that we have a reference file right of your existing newsletter

[译文] [Amir]: 好的。所以我想——在这种情况下我会做的是确保我们有一份你现有 Newsletter 的参考文件，对吧？

[原文] [Greg Isenberg]: I would need a and would you need like an export of all my tweets Exactly Yeah Yeah So maybe we can try to do one example one right now and then um and then see

[译文] [Greg Isenberg]: 我需要——你会需要导出我所有的推文吗？

[Amir]: 没错，是的，是的。

[Greg Isenberg]: 所以也许我们可以现在先试做一个例子，然后再看。

[原文] [Amir]: So what's a good tweet Yeah You find a tweet that like speaks to you and then we'll Okay I like this one actually It speaks to me Okay Pricing

[译文] [Amir]: 那么哪条推文比较好？是的，你找一条能打动你的推文，然后我们——好的，我其实喜欢这条。它打动了我。好的，关于定价的。

[原文] [Greg Isenberg]: Yeah Yeah I mean this is a perfect one cuz it was too long for Twitter Yeah But I'd still posted it anyways Mhm Um but if I did even if I did this on a newsletter I would totally expand on this

[译文] [Greg Isenberg]: 是的，是的。我的意思是这是一个完美的例子，因为它对 Twitter 来说太长了。但我还是发了。嗯，但如果我真的把它放在 Newsletter 里，我会完全扩展它。

[原文] [Amir]: All right cool So let's take this actually and we're going to copy this and then create notes Export this example newsletter Export it as a markdown Right Yeah As a markdown Exactly Exactly So we're going to go back to Claude now So now that we have an example tweet post and example newsletter um we're going to um up uh we're going to upload that as a reference file in there as well

[译文] [Amir]: 好的，酷。所以我们选这个，复制它，然后创建笔记。导出这个示例 Newsletter。导出为 Markdown，对吧？

[Greg Isenberg]: 是的，作为 Markdown。

[Amir]: 没错，没错。所以我们现在回到 Claude。既然我们现在有一篇示例推文和一篇示例 Newsletter，我们要把那个也作为参考文件上传进去。

[原文] [Amir]: So I have just had the skill Can you turn this tweet into a newsletter format And you you're going to be the judge of this Tell me if you think it's good Yeah

[译文] [Amir]: 所以我已经添加了这个 Skill。（输入指令）“你能把这条推文变成 Newsletter 格式吗？”你来当裁判。告诉我你觉得它好不好。是的。

[原文] [Amir]: So honestly this is fire I mean so tone of voice So we just did this in one shot but if we really wanted to we would take all of your existing tweets all of your newsletters and then use that to generate a like a style guide or tone of voice and then kind of refine this But as a starting point it's not bad As a starting point it's not bad And I I I'll even take it a step further Like this is It's actually not bad Really good It's actually not bad at all

[译文] [Amir]: 老实说，这真是太火了（Fire）。我是说这种语调。我们只是一次性做的尝试，但如果我们真的想做，我们会把你所有的现有推文、所有的 Newsletter 都拿来，然后用它们生成一个风格指南或语调指南，然后再进行微调。但作为一个起点，这还不错。作为一个起点，这真的不错。我甚至要更进一步说——这实际上还不错。真的很好。这实际上一点也不差。

[原文] [Greg Isenberg]: Until next time keep building and keep raising Not keep raising Yeah We don't you know but keep building I like uh forward this to a founder who's been sitting on the same price point for too long Like I like that I think that's really smart Right We talking about product market fit We should talk more about price pricing market fit Like that's a that's a banger That's a banger That's actually a banger

[译文] [Greg Isenberg]: （读生成的内容）“直到下一次，继续建设，继续融资（raising）……”不，不是继续融资。是的，我们不——你知道，但是“继续建设”。我喜欢这句：“把这个转发给那些在同一定价点上停留太久的创始人。”比如我喜欢这个，我觉得这真的很聪明。对吧？我们谈论产品市场契合度（Product Market Fit），我们应该更多地谈论定价市场契合度（Pricing Market Fit）。这简直是个杰作（Banger）。这是个杰作。这真的是个杰作。

[原文] [Amir]: Actually yeah Here's where it gets interesting though right Cuz you can now create skills that generate visual graphics cuz that's that's a thing now You can so you can you don't have to do MCP calls like Canva or anything like that You can programmatically create these visuals So we can update the skill to say "Hey now add images as well in there."

[译文] [Amir]: 确实是的。不过这里有个有趣的地方，对吧？因为你现在可以创建生成视觉图形的 Skills，因为现在这也是个功能了。你可以——所以你不必进行像 Canva 之类的 MCP 调用。你可以通过编程方式创建这些视觉效果。所以我们可以更新这个 Skill 说：“嘿，现在在里面也加上图片。”

[原文] [Amir]: So and we've covered some use cases as well The last thing I want to talk about is kind of just like you know I saw this report I don't know if you if you saw from ramp where they were tracking subscriptions for like different AI tools and they saw that there's a dip happening and they're saying there's getting stickier in enterprise but the AI is off ramping and it's not as sticky as we want it to be because cost is coming down

[译文] [Amir]: 所以我们也涵盖了一些用例。我想谈的最后一件事就像是——你知道我看到一份报告，不知道你是否看到了 Ramp 的报告，他们在追踪不同 AI 工具的订阅情况，他们看到出现了一个下降（dip）。他们说企业端的粘性在增加，但 AI 正在被“下架（off-ramping）”，它没有我们希望的那么有粘性，因为成本在下降。

[原文] [Amir]: I actually want to say now with what we're seeing with skills and all the education and awareness around prompting we should be able to solve that gap because the reality is a lot of companies are investing in AI and there's reports now saying that it's not actually being as productive as we thought The issue is I think prompting and context the issue is people The issue is not Yeah Yeah That's the reality right There isn't enough AI fluency and education around how to actually do prompting

[译文] [Amir]: 我实际上想说，随着我们看到的 Skills 以及围绕提示词的所有教育和意识的提升，我们应该能够解决这个差距。因为现实是，很多公司都在投资 AI，而现在的报告说它实际上并没有像我们想象的那样富有成效。我认为问题在于提示词（Prompting）和上下文，问题在于人。问题不在于（AI本身）。

[Greg Isenberg]: 是的，是的。

[Amir]: 这就是现实，对吧？关于如何真正进行提示词操作，缺乏足够的“AI 流畅度（AI Fluency）”和教育。

[原文] [Amir]: People like write you know build me a SAS 1 million AR don't make mistakes you know and the reality is like no no you got to give the right amount of context and do some prompt structure And um I think when it comes to anthropic they do a really good job of not only building with intention but creating the resources the education to help people actually become more AI fluent and giving them tools to do that

[译文] [Amir]: 人们喜欢写这种：“给我建一个年收入 100 万美元的 SaaS，别犯错”，你知道吗？而现实是，不，不，你必须给出适量的上下文并做一些提示词结构。而且我认为谈到 Anthropic，他们真的做得很好，不仅仅是有意图地构建产品，还创造了资源和教育来帮助人们真正变得对 AI 更熟练，并给他们工具来做到这一点。

[原文] [Amir]: So um the net of this takeaway is AI adoption may be falling and the adoption rates may be down for this month or this past quarter I think part of it is just because companies don't have the right resources and the people to build education um on AI enablement and AI fluency and then once we see that come into play adoption is going to come back up and there's the tools now to support that

[译文] [Amir]: 所以这个结论的核心是，AI 采用率可能在下降，这个月或上个季度的采用率可能下降了。我认为部分原因只是因为公司没有正确的资源和人员来建立关于 AI 赋能和 AI 流畅度的教育。一旦我们看到这一点发挥作用，采用率就会回升，而且现在有工具可以支持这一点。

[原文] [Greg Isenberg]: Beautiful Well thanks for explaining it to me honestly and everyone else Um for more of air uh I'll include links in the show notes where you can go ahead and follow him Uh X is the best place Yeah Air MXT A M I R MXT Cool I appreciate you coming on Cool Thanks for having me Thanks man Thanks

[译文] [Greg Isenberg]: 太棒了。老实说，谢谢你向我和大家解释这一切。想了解更多关于 Amir 的信息，我会把链接放在节目简介里，大家可以去关注他。X（推特）是最好的地方吗？

[Amir]: 是的，Amir MXT。A-M-I-R-M-X-T。

[Greg Isenberg]: 酷，很感激你能来。

[Amir]: 酷，谢谢邀请我。

[Greg Isenberg]: 谢谢兄弟，谢了。