适用于产品负责人的Claude Code：研究、写作、上下文库、自定义待办系统等 | 特蕾莎·托雷斯

为什么克劳德代码成了你的伙伴？

Why has Claude Code become your buddy?

我以前用任务管理工具记笔记，那时候用的是Trello。

I was writing my notes in my task management tool, and that was Trello.

随着时间推移，我开始非常担心，我到底该怎么把数据从Trello里导出来。

As time went on, I just started to get really worried about how am I ever gonna get my data out of Trello.

于是我心想，不知道克劳德能不能帮上忙。

And so I was like, I wonder if Claude can help.

那只是其中一件事。

And that was like one thing.

也许我可以用克劳德把这件事做得更好。

Maybe I could just do this better with Claude.

当我把任务管理移到克劳德后，现在克劳德能看到我的任务，我每天早上可以直接说：克劳德，你帮我列一下今天该做的事，哪些你能直接帮我完成？

And by moving my task management to Claude, now Claude sees my tasks and I can literally start my day and be like, Claude, what's on my to do list that you can just do for me?

我还可以问：嘿，克劳德，我现在的销售漏斗情况怎么样？

I can say, Hey Claude, what's my sales pipeline right now?

由于克劳德会为我的任务打标签，它能自动生成我所有销售任务及其当前状态的列表。

And because Claude is tagging my tasks, it literally can generate a list of all my sales tasks and where they're at.

这给了我很多灵感，因为我经常忘记很多事情。

This has given me a lot of inspiration because I forget a lot of things.

事情太多了。

There's a lot going on.

欢迎回到《我是如何使用AI的》。

Welcome back to How I AI.

我是克莱尔·沃，产品负责人兼AI狂热者，致力于帮助你更好地使用这些新工具。

I'm Claire Vo, Product Leader and AI Obsessive here on a mission to help you build better with these new tools.

今天，我们邀请到了《持续发现习惯》的作者特蕾莎·托雷斯，这本书我们大家都熟知并喜爱，她是一位享誉国际的作者、演讲者和教练。

Today, we have a very practical episode with Teresa Torres, author of Continuous Discovery Habits, which we all know and love, an internationally acclaimed author, speaker, and coach.

特蕾莎将向我们展示她是如何用克劳德代码来处理几乎所有事情的，尤其是管理她庞大的待办事项清单、所需的所有信息，并构建一个庞大的上下文库，以便她能像自己说的那样，用更轻松的方式进行提示。

Teresa's gonna show us how she uses Claude Code for basically everything, but especially to manage her huge to do list, all the information she needs to do a great job, and builds a giant context library so she can be, as she says, lazy with her prompting.

我们开始吧。

Let's get to it.

本集由Brex赞助。

This episode is brought to you by Brex.

如果你正在收听本节目，你已经知道AI正在以真实而实际的方式改变我们的工作方式。

If you're listening to the show, you already know AI is changing how we work in real practical ways.

Brex正将同样的力量引入财务领域。

Brex is bringing that same power to finance.

Brex是一个专为创始人打造的智能财务平台。

Brex is intelligent finance platform built for founders.

通过后台运行的自主代理，你的财务系统几乎可以自动运行。

With autonomous agents running in the background, your finance stack basically runs itself.

卡片被发放、支出被报销、欺诈行为在实时中被阻止，而你无需操心。

Cards are issues, expenses are filed, and fraud is stopped in real time without you having to think about it.

将Brex的银行解决方案与高收益现金管理账户结合，你就拥有了一个帮助你更明智地支出、更快地行动、自信扩张的系统。

Add Brex's banking solution with a high yield treasury account, and you've got a system that helps you spend smarter, move faster, and scale with confidence.

美国每三家初创企业中就有一家正在使用Brex。

One in three startups in The US already runs on Brex.

你也可以在 brex.com/howiai 上体验。

You can too at brex.com/howiai.

特蕾莎，欢迎来到《How I AI》。

Teresa, welcome to How I AI.

我非常期待今天的节目，因为我们能见证我所热爱的——

I am so thrilled with today's episode because we get to see what I love, which

在非编程环境中使用经典的 Claude Code。

is good old Claude Code in a non coding environment.

我们在节目开始前还笑了一会儿。

And we were laughing before the show.

你刚刚在电脑的每个目录里，直接在终端里操作。

You were just in every directory on your computer straight in the terminal.

这就是我们现在的日常生活方式。

Like, this is how we're living right now.

所以在我们深入之前，我必须问你，为什么选择 Claude Code？

So I have to ask you before we get into any of it, why Claude Code?

为什么克劳德代码成了你的伙伴？

Why has Claude Code been become your buddy?

是的。

Yeah.

这其实是一个非常渐进的过程。

This has been a really gradual evolution.

我刚开始和大家一样，纯粹在网页版的ChatGPT上使用。

I started like everybody else, like literally in ChatGPT in the web.

后来我想找个大语言模型帮我写作，就慢慢转到了克劳德，因为克劳德的写作能力稍好一些，虽然这一点可能正在改变。

And then I wanted to have an LLM help me with writing and I gradually moved to Claude because Claude's a little bit better writer, although that might be changing.

然后，我确实会写代码，主要在AWS环境中编程，这可能会让我有点尴尬。

And then, you know, I do code and I mostly code in the AWS environment and this is going be really embarrassing.

我大部分编码工作都是在AWS管理控制台里完成的。

I do most of my coding in the AWS management console.

我丈夫四年来一直跟我说：特蕾莎，你真该用个IDE了。

And my husband for like four years has been like, Teresa, you just need to use an IDE.

我之前有点害怕用IDE。

And I was kind of afraid of IDE.

我根本就没有使用版本控制。

I literally had no version control.

但最近我参与了一个项目，我开发的一个功能被集成到了一个真正的生产级产品中。

But I recently had a project where one of the things that I built is being integrated into a real production quality product.

我当时就想，我得提升我的工程水平了。

And I was like, oh, I got to level up my engineering game.

这让我开始使用VS Code。

And that got me into VS Code.

然后，因为如果我想让我开发的东西进入真实产品，我就得使用Git，像个真正的工程师一样。

And then because I needed to use like Git and be like a real engineer if I was going to have something I built go in a real product.

所以我不得不提升我这‘假装的’工程技能，我丈夫说：你看，你直接在VS Code里就能让Claude在终端里帮你。

And so, I had to level up my pretend engineering game and my husband was like, And look, you can just have Claude in the terminal right here inside VS Code.

那之后，一切都改变了。

And then that was a game changer.

我认为使用Claude进行编码帮助我实现了我们今天要讨论的所有事情，因为我感觉工程师在使用Claude Code时，就像是和Claude一起结对编程。

And I think the reason why coding with Claude helped me with all the things we're going to talk about today is I feel like engineers pair program with Claude when they use Claude Code.

我认为结对编程这个理念，现在我已经应用到我做的每一件事上，即使不是编程也一样。

And I think this idea of pair programming, I pair program now with everything I do, even if it's not programming.

所以我不仅结对管理任务，还结对写作，结对做所有事情。

So like I pair task manage and I pair write and I pair everything.

太棒了。

Great.

那我们开始吧，因为你将为我们展示任务管理。

And so let's get into it because you're gonna show us task management.

有个老笑话是说，每三家初创公司里就有一家是做待办事项清单的。

And I mean, running joke is like every third startup is going to be a to do list.

如果你没试着创办过一个待办事项清单的公司，那你还算得上是真正的早期创始人吗？

Like, if you haven't tried to start a to do list startup, are you really like an early stage founder?

但你为自己编码了一个适合你的任务管理平台，我很想请你为我们详细介绍一下。

But you have coded yourself a task management platform that works for you, and I'd love for you to walk us through it.

是的

Yeah.

我会解释我为什么这么做。

And I'll explain why I did this.

我实际上认为，任务管理应用如此之多的原因在于，我们管理任务的方式非常个性化，这正是你应该为自己打造工具的类型，因为它可以完全按照你的需求运作，这也是其魔力所在。

I actually think the reason why there's so many task management apps is that how we manage our tasks is so idiosyncratic that this is exactly the type of thing that you should build for yourself because it can work exactly the way you want it to work, which is part of the magic.

但让我走到这一步的原因是我是个重度笔记爱好者。

But what got me here was I'm a huge notetaker.

与其大声思考，我更倾向于通过列提纲和写笔记来思考。

Like, instead of thinking out loud, I think by outlining and writing notes.

我当时把笔记写在任务管理工具里，那个工具是Trello。

And I was writing my notes in my task management tool and that was Trello.

这简直是个噩梦，因为我的笔记现在被锁定在第三方工具里了。

And this is kind of nightmarish because now my notes are locked into a third party tool.

而且这些笔记非常难搜索。

They weren't very searchable.

随着时间推移，我开始非常担心，我究竟要怎么才能把数据从Trello里导出来。

And like, as time went on, I just started to get really worried about how am I ever going to get my data out of Trello.

于是我心想，不知道Claude能不能帮上忙。

And so I was like, I wonder if Claude can help.

那是一方面，也许我可以用Claude把这件事做得更好。

And that was like one thing, like maybe I could just do this better with Claude.

而让我深入这个领域的原因还有另一方面：当我越来越投入AI时，我强迫自己每次做任务时都问：AI能怎么帮助我？

And then the second thing that got me into this was as I started to get more involved in AI, I forced myself every time I did a task to ask, how can AI help with this?

它能自动化这个任务吗？

Can it automate it?

它能增强这个任务吗？

Can it augment it?

我喜欢做这个吗？

Do I like doing it?

我想要AI替我做吗？

Do I want AI to do it for me?

通过将我的任务管理交给Claude，现在Claude能看到我的任务，我每天早上可以直接说：Claude，你有哪些任务可以替我完成？

And by moving my task management to Claude, now Claude sees my tasks and I can literally start my day and be like, Claude, what's on my to do list that you can just do for me?

或者，有哪些任务是我应该思考如何让你帮忙的？

Or what's on my to do list where I should be thinking about how you can help?

这样一来，我就不用再费心去琢磨怎么使用AI了。

And then it's not on me to figure out how to use AI.

Claude就像是我的AI搭档。

Claude's kind of my pair of AI buddy.

所以我做了一个名为/今天的斜杠命令。

So what I've built is I have a slash command slash today.

大家熟悉斜杠命令吗？

Are people familiar with slash commands?

我要不要解释一下？

Should I describe that?

你 definitely 应该解释一下。

You should definitely describe them.

好的。

Okay.

斜杠命令就是我们自定义的Claude Code快捷方式。

So a slash command is just a Claude Code shortcut that we get to define.

所以我决定把它命名为斜杠today。

So I decided it was called slash today.

我写了一个非常详细的提示词，我们可以看一下，它会告诉Claude每次我输入斜杠today时该做什么。

I wrote a really detailed prompt which we can look at that tells Claude exactly what to do every time I type slash today.

所以每天早上，包括周六日，我一边喝着咖啡，一边坐下，直接输入斜杠today。

So every single morning of my life, even on Saturdays and Sundays with my cup of coffee, I sit down and I literally just type in slash today.

如果我现在运行它，会覆盖我待会儿要给你们看的内容，但我们稍后会实时运行一次。

And if I run this, it's going to overwrite what I'm going to show you but we'll run it real live in a minute.

但你可以看到今天早上输出的摘要内容。

But what it does, you can see the summary from this morning's output.

它会检查我的Trello看板，因为我仍然用Trello来和团队协调。

It checks my Trello board, so I still use Trello to coordinate with my team.

它说没有新卡片，我的列表里没有什么新内容要添加。

Says there were no new cards, nothing new to add to my list.

它生成了一个今天的文件，这就是我们在Obsidian里看到的这个文件。

It generates a today file, which is what we're looking at here in Obsidian.

它已经检查了我所有的任务，这些任务都是Markdown文件，并告诉我今天有哪些截止任务。

This is it's gone through all of my tasks that are just markdown files and told me what's due today.

现在时间已经比较晚了，我已经完成了很多待办事项。

We're late in the day so I've done a lot of my to do list.

我和其他人一样。

I am like every other human.

我有一长串 overdue 的任务还没完成，它们总是排在我的待办列表最前面。

I have a long list of overdue tasks that I have not done and they always end up at the top of my to do list.

任务就是那些有截止日期的事情，对吧？

And tasks are things that have due dates, right?

比如一些必须在特定时间之前完成的事情。

Like they're things I have to do by a certain time.

我还有一个文件夹，里面全是各种想法，都是我哪天想做的事情。

I also have a whole folder full of ideas, just things that I want to get to someday.

你可以看到，我目前有四个想法正在进行中。

And you can see I have four ideas that are currently in progress.

这些想法每天都会被添加到我的待办清单里，这样当我做完清单上的其他任务时，就可以想想：接下来该专注哪些长期项目？

These get added to my to do list every day so that when I make it to the end of my list, can be like, Okay, what should I be working on that's more long term?

接下来我们可能会谈到这个，但我确实创建了一个插件，每天帮我执行研究查询。

And then we're going to talk about this, I think next, but I do have this kind of plugin that I created that does research queries for me every day.

每天在我的待办清单里，我都会收到一份研究摘要，用来查阅、保存那些我想总结和学习的论文。

And then every day on my to do list, I get, my research digest to review and save papers that I want to, summarize and learn from.

这个工作流看似简单，但背后的机制是这样的：如果我往左边看，我只有几个不同的文件夹。

And that's it's such a simple workflow, but what's behind the scenes, if I go over here on the left, I just have different folders.

这些文件本质上都是Markdown文件。

These are literally markdown files.

我们正在使用Obsidian。

We're in Obsidian.

当我有一个错误文件夹、一个想法文件夹和一个任务文件夹时。

When I have a bugs folder, I have an ideas folder, I have a tasks folder.

如果我们看一下一个任务，任务会有一些前端信息。

And then if we look at a task, a task has some front matter.

前端信息是Obsidian的一个概念。

So front matter is an Obsidian concept.

它就是字段类型加上对应的值。

It's just like field type and then value.

底层实际上是YAML格式，对熟悉它的人来说就是这样。

It's like, it's YAML behind the scenes for people familiar with that.

所以我有任务类型，任务有截止日期，还有标签。

And so I have a type tasks, task has a due date, and then there's tags.

我的每一个任务都包含这些信息。

And every single one of my tasks has this.

所以当我运行我的今日命令时，Claude只是在我的任务文件夹中搜索所有截止日期为今天的项目。

And so what's happening when I run my today command is Claude is just searching my tasks folder for anything that has a due date of today.

我想问一下，后台是不是通过 Atlassian MCP 拉取 Trello 的数据？

And I have to ask just behind the scenes, is the Trello data being pulled via the Atlassian MCP?

还是说它是如何实际访问所有这些数据的？

Or how is it actually accessing all this data?

Obsidian 是本地存储的吗？

Is Obsidian stored locally?

这些功能到底是怎么整合在一起的？

Like, how does this all stitch together?

是的。

Yeah.

实际上，我再也不用 Trello 了。

So I actually don't use Trello anymore at all.

我来告诉你为什么我还有一个 Trello MCP 服务器，但现在当我想要创建任务时，我不会再跑去 Trello 看板上创建了。

I'll tell you why I have a Trello MCP server, but I don't Like, now when I wanna create a task, I don't go create it on my Trello board.

我可以现场演示一下。

I literally We can demo this.

新建任务。

New task.

给克莱尔发个感谢。

Send thank you to Claire.

真贴心。

Very sweet.

太棒了。

Was a blast.

然后，Claude 拥有我任务管理系统的所有上下文，因为 Claude 是在我任务文件夹中打开的。

And then Claude has has all the context for how my task management system works because Claude is open in my tasks folder.

你可以看到，它正在我的任务文件夹中创建一个文件。

And you can see here, it's creating a file in my task.

明白了。

Got it.

它把截止日期设为今天。

It set the due date to today.

它没有添加标签，这有点问题。

It hasn't added tags which is kind of a problem.

我们看看它能不能自己弄明白。

We'll see if it figures it out.

哦，它真的很聪明。

Oh, it's been very smart.

因为它把我的一项任务加到了今天的列表里，所以它知道需要更新我们正在看的这个今天的文件。

So because it added something to my today list, it knows it needs to update my today this file that we're looking at.

我其实不希望它这样处理，因为这会删除我所有的清单，而我喜欢有完成任务的成就感。

I actually don't want it to do it this way because it's gonna remove all my checklists and I like feeling like I did my stuff.

所以我干脆直接告诉它手动添加吧。

I'm So gonna just tell it to just manually add it.

这就是它刚刚尝试运行的脚本，向你展示了这个今天/命令背后的运作机制。

So that's the script that it just tried to run as kind of telling you the behind the scenes of how this today slash command works.

背后其实有一个Python脚本在执行这些操作。

There is a Python script behind this that like does that.

它会搜索所有任务，查找任何今天到期或已过期的内容。

It's that does that like search all the tasks for anything that is due today, search for anything that's past due.

你可以看到，现在它直接出现在我的待办事项列表中了。

And you can see here that like now it just shows up on my to do list.

这听起来可能很傻，但我根本不需要打开网页浏览器。

And this sounds so silly, but I didn't have to open a web browser.

我不需要在不断变化的图形界面里点击十几个按钮。

I didn't have to click through 14 different buttons in a GUI that is constantly changing.

我不需要点日期选择器，再点标签，然后把它移到正确的列表里。

I didn't have to, like, click on a date picker and then click a label and then move it to the right list.

我其实就是随手打了几行笔记发给克劳德。

Like, I literally just typed, like, off the cuff notes to Claude.

因为我整天都用克劳德，这个任务窗口一直开着，我通常还会开另一个会话来处理当前项目。

And because I work in Claude all day every day, this task window is always open and then I usually have a second session open for whatever project I'm working on.

所以我随时可以切换过去，说一声‘新任务’或者‘新想法’。

And so I can always just bounce over and be like, hey, new task or hey, new idea.

我最喜欢的就是它的速度。

And it's just it's it's the speed of it is what I really love about it.

所以我什么都不用想。

So I don't have to think about anything.

那为什么要把Obsidian加入流程呢？

And then why put Obsidian in the loop?

这本来就是你已经有的东西。

It's something you already had.

它包含了很多上下文信息。

It has a lot of context.

它的结构正是你想要的样子。

It's structured the way you want it.

这些本来就可以是纯markdown文件，你直接像上面那样打勾就行。

You know, these could be just raw markdown files, and you could just x them off the way they're shown above.

你觉得这个额外的层次对你来说有什么意义？

What do you think that extra layer is for you?

在这之前我并不是Obsidian的用户，而且一开始我对markdown也不太熟悉。

I was not an Obsidian user before this, and I'll say I wasn't even comfortable with markdown at the beginning.

让我们回溯到六个月或八个月前。

Like, let's go back six, eight months.

那时我根本不是一个熟练使用markdown的人。

Like, I was not a comfortable markdown user.

我有几样喜欢的东西。

I there's a few things I like.

我喜欢那种非常直观的体验，虽然听起来有点傻，但我就是喜欢打勾。

I like the really tactile, like, it's silly, but I like checking the box.

我知道我把它放那儿了。

I know I was put it

我正要问。

gonna ask.

是的。

Yeah.

我知道我可以在方框里打个x，用markdown。

I know I can put an x in a box and markdown.

但感觉不太一样。

It's not quite the same.

我真正喜欢的是左边的文件浏览器。

Really what I like is the file browser on the left.

如果你注意一下，我的资料库并没有只设为任务，而是更高层级的。

And if you'll notice, my vault is not set at tasks, it's set higher than that.

所以我有——我们待会儿聊聊我的LLM上下文，但我把我的LLM上下文、所有笔记、一些用于处理播客的音频文件都放在一起。

So I have and we'll talk about my LLM context, but I have like my LLM context, all my notes across everything, some podcast files that I use to work with my podcast.

这些都是我们要深入探讨的研究内容，我的所有任务、一些我正在尝试的技能，还有我的写作。

This is the research stuff we're gonna get into, all my tasks, some skills that I've been trying to experiment with, my writing.

所以我把Obsidian看作是我的文件浏览器。

And so I kind of think about Obsidian as my file browser.

而且因为所有内容都是markdown格式，这让我做的每一件事都能被Claude轻松访问。

And because it's all in markdown, it makes everything I do super accessible to Claude.

然后我可以做这样的事，比如对Claude说：‘嘿，Claude，我现在的销售漏斗怎么样？’

And then I can do things like I can say, Hey, Claude, what's my sales pipeline right now?

因为Claude会为我的任务打标签，它能实时生成我所有销售任务的列表及其当前状态。

And because Claude is tagging my tasks, it literally can generate a list of all my sales tasks and where they're at on the fly.

对于大多数任务管理工具来说，你只能使用它们提供的视图或手动打标签，但谁会真的手动打标签呢？

So like for most task management, you're limited to what views they create or you can use tags, but who manually tags things?

我本以为自己会这么做，但我从来都没做过。

I'm optimistic I'll do that, but I never do it.

而在这个系统中，所有标签都由Claude来完成。

Whereas in this system, Claude does all the tagging.

每次它生成任务时，都会自动思考该添加哪些标签。

Anytime it generates a task, it'll think about what tags to add.

在本项目的Claude MD文件中——不是我的全局文件，而是这个项目专用的——我们会维护一个标签体系，并加以管理。

And then in my Claude MD for this project, not for my global one, for this project, we keep a taxonomy of what tags we're using and we kind of manage that.

当我看到不喜欢的地方，就会更新这个Claude MD文件，这样我和Claude是在共同创建它，但所有繁重的工作都是Claude在完成。

When I see things I don't like, I update that Claude MD so that, like, I'm co creating it with Claude, but Claude's doing all the heavy lifting.

在我们进入下一个工作流之前，我想先总结一下：你已经在Claude Code中创建了一个斜杠命令，用于查看你的任务并整合你今天的任务。

So I want to recap this before we go to the next workflow, which is you've created a slash command in Claude Code to look at your tasks and assemble basically assemble your tasks from today.

你在Obsidian中使用了一种结构化的任务文档格式，每个任务都会被录入，包含标题、截止日期、自动填充的标签以及一些上下文信息。

You have a structured task document format in Obsidian that every task goes into with like a title, a due date, some tags that automatically get populated and some context.

此外，你还定义了一些可以在Claude中使用的固定命令，能够实时添加、删除或更新这些任务。

And then you have a couple like known commands you can use in Claude that allow you to add, remove, update, whatever those tasks on the fly.

这为你提供了一个个性化的待办事项体验，能够跨所有来源无缝连接。

And this just gives you the personalized experience you want for your to do list connected across all your sources.

如果有人忽略了这一点，我想特别指出：你提到Claude可以设置项目级指令和全局级指令。

And then in case it slipped by people, do wanna call out, you noted that Claude can have sort of like project level instructions and global level instructions.

我认为这一点很多人没有充分利用，那就是将你的Claude MD文件根据上下文范围进行精准配置，比如为你的任务管理列表专门设置一个专注的版本。

And I think this is something that people don't take enough advantage of is context scoping their clot MD files to the right area so that you could have one for your task management list that's really focused.

而你还有一个全局版本，用于管理所有属性等。

You have a global one for all your properties, etcetera, etcetera.

所以，我想我理解你的整个流程了。

So I think I think I have your flow.

这给了我很多灵感，因为我经常忘记很多事情。

This has given me a lot of inspiration because I just forget a lot of things.

事情太多了，好吧。

There's a lot going Okay.

我来给你展示一下真正的价值。

I'm gonna show you like, here's the real value.

比如说，我现在在做这个任务。

Like, let's say I'm doing this task.

这个任务并不简单。

This isn't too simple of a task.

比如说，我正在筹备一门课程，对吧？

Let's do, I'm working on launching a course, right?

当我开始这个任务时，假设我之前刚完成销售页面的一半更新。

And if- as I get onto this, like, let's say I'm like, right before this, I was like, half done updating the sales page.

然后假设我在课程平台上发现了一个bug，需要记录下来。

And then let's say I find a bug in my course platform and I need to document this bug.

我在做任务时会直接把所有笔记记下来，而且都嵌入在其中。

I take all my notes literally while I'm doing the task and it's all embedded.

而且，这些全部都是文字内容。

And again, it's all text.

所以，如果以后，比如明天我回来时想：‘我到底把那个bug记哪儿了？’

So if later, like tomorrow I come back and be like, Where in the world did I log that bug?

因为我只是懒散地记在笔记里，并没有真正创建一个bug记录。

Because I did it lazily in my notes and I didn't actually create a bug.

我就可以说：‘Claude，帮我找找这个我都不记得在哪的东西。’

I can be like, Claude, help me find this thing that I don't know where it's at.

我不确定你是不是这样。

Like, I don't know about you.

我不知道你有没有用过Trello，但我觉得任何我可能提到的任务管理工具，其搜索功能都不太好，也无法很好地搜索任务中的所有上下文信息。

I don't know if you've used Trello, but I think any task management tool that I could say this about, the search is not that good, and it's not that good at searching all the context in the task.

而我特别喜欢这一点：即使我自己找不到，Claude也会尝试各种搜索组合，直到找到它。

And that's what I really love from this is that like, I can't figure it out, but Claude will try every permutation of searches till it finds it.

即使我记错了词语，我也会说：嘿，我明天有个叫‘新博客文章’的东西。

Even if I'm remembering the words wrong, I'll be like, hey, I have a thing called new blog post tomorrow.

然后它会说：我找不到叫‘新博客文章明天’的东西，但我有个叫‘文章周三’的内容。

And it'll be like, I can't find anything called new blog post tomorrow, but I have this thing that says article Wednesday.

你找的是这个吗？

Is that what you're looking for?

我会说：哇，Claude。

And I'll be like, woah, Claude.

这正是我找的东西。

That is what I'm looking for.

是的。

Yeah.

我觉得我们还没充分意识到，这些其实是极其出色的本地搜索引擎，毕竟我们在写代码时经常用到。

I don't think we say it enough that these are really great local search engines for for I mean, we use it so much in code all the time.

我说，嘿。

I'm like, hey.

你能提醒我一下 x y z 是怎么工作的吗？

Can you remind me how x y z worked?

或者我觉得我已经上线了这个功能。

Or I think I shipped this feature.

你能具体提醒我它是怎么实现的，或者是谁做的吗？

Can you remind me exactly how it was implemented or who did it?

但你可以将同样的框架、同样的搜索方式应用到这些基于文本的工具中。

But you can apply that same framework, that same like search framework to kind of any text based tool in these.

这些工具在捕捉上下文方面非常出色。

These tools are really good at grabbing that context.

说到寻找有用的上下文，你还有一个关于研究的流程，我很想请你详细讲讲。

Well, speaking of finding useful context, you have a second workflow around research I would love for you to walk us through.

那么，你是如何整理这些帮助你工作的研究资料的呢？

So how do you assemble all this research that helps you do your job?

好的。

Okay.

我立志成为一名学者，这听起来有点奇怪，但我确实这么想。

So I aspire to be an academic, which is weird, but I do.

我非常希望跟上各个领域的学术研究。

And I really want to keep up on academic research on a lot of topics.

事实上，我们可以来看看我的研究主题。

In fact, we can go over here and look at my topics.

比如，我特别关注关于合成用户的研究，我们是否应该让AI来为我们完成访谈整合？

So like, I do a lot about I'm really interested in this research around like synthetic users and should we be letting AI do interview synthesis for us?

我对团队协作、创造力、探索能力等话题都很感兴趣。

I'm interested in team collaboration, creativity, discovery skills, whatever.

还有教育，因为我教书；还有用户画像，因为这是个非常热门的话题。

Education because I teach, personas because it's a super hot topic.

我真的很想知道，学术研究正在教给我们什么？

And I really want to know what are we learning from academic research?

我恰好能访问大学图书馆，这很棒，但我总是缺乏主动去搜索资料的自律。

I happen to have access to a university library, which is great, but I never have the discipline to like go search for things.

我一天中从来没有过觉得无聊的时刻。

There's never a moment in my day where I'm like, Oh, I'm bored.

我应该去做的。

I should go do this.

但我真希望我能去做，对吧？

But I wish that I did, right?

所以我做了这么一件事，这也是使用Claude作为我的任务管理器的一个好处——我可以直接把它集成到我的任务管理器中。

And so what I did, and this is also one of the nice things about using Claude as my task manager is I can integrate it right into my task manager.

所以我会先给你看看这个的输出结果，然后再讲我是怎么搭建它的。

So I'm going to start by showing you the output of this, and then I'll talk about how I built it.

所以每天在我的待办事项列表里，我都会收到一份简短的研究摘要。

So every day on my to do list, I get a little research digest.

它会给我提供每日存档搜索的结果。

It's giving me the search results from a daily archive search.

Archive是一个预印本服务器。

So archive is a preprint server.

现在大多数论文，得益于新冠疫情及其后的影响，都是在正式发表前先发布在这里的。

It's where most papers now, thanks to COVID and post COVID, get published before they're published for real.

好处是这些论文免费且更新迅速。

What's nice is they're free and they're fast.

它们是实时的。

They're real time.

不好的地方是，这些论文尚未经过编辑，因此你需要自己充当筛选者。

What's not nice is it's before they're edited, so you have to be your own filter.

它会进行搜索，然后我收到一个包含所有结果的 Markdown 文件。

It does a search and then I get a markdown file with all the results.

所以这个文件就是今天生成的。

So this is like, this is literally today's file.

我还没来得及看，但当我查看时，如果打开并下载了某个 PDF，我会把它保存到相应的主题文件夹中，而每个主题文件夹下都有一个 sources 目录和一个 notes 目录。

I have not gone through it yet, but then when I go through it, if I open a PDF and I download it, I save it to these topic folders, And in each of these topic folders, there's a source directory and a notes directory.

所以我的 PDF 会放在 sources 里，这一点稍后会很重要。

So my PDF goes in sources, and this is going to matter in a second.

当我保存完一篇PDF后，第二天在《今日研究摘要》中，我会收到前一天保存的每篇论文的摘要。

And then, what happens the next day after I've saved a PDF is on this Research Today Digest, I get summaries of every paper I saved the day before.

我会得到非常详细的摘要，不是那种半成品的、仅有一段话概述论文内容的摘要，而是我专门编写了这个功能来聚焦于论文的方法和效应量，因为作为编辑，我必须自己判断这些关键点。

And I get really detailed summaries of like not kind of the half baked, here's a paragraph of what the paper is about, but I wrote this skill to like very to focus on like the methods of the paper and the effect size, things that like are going to because I have to be the editor.

这些论文目前还没有编辑。

There's no editor for these papers yet.

这些信息能帮助我决定：这篇论文值得读吗？

Things that can help me decide, is this worth reading?

它的效应量是否足够大？

Does it look like it had a big enough effect size?

它看起来像一项严谨的研究吗？

Did it look like it was a good study?

关于这一点，我有个有趣的故事。

And I have a funny story about this.

我开发这个功能的当天，我在查看每日摘要时，看到了一篇关于购买意向的论文。

The day I built this, I was reviewing my daily digest and I saw this paper on purchase intent.

我读了这篇论文，它被总结了，我看了摘要，因为摘要格式很清晰，我发现了这个缺陷。

And I read this and I had it summarized and I read the summary and because it was like in this nice summary format, I realized this flaw.

我当时想，哦，他们用的购买意向调查并不可靠。

I was like, Oh, they use this purchase intent survey that's not very reliable.

这并不是衡量购买意向的准确方法。

Like it's not an accurate measure of purchase intent.

第二天在LinkedIn上，我看到Ethan Moloch分享了这篇论文。

And the next day on LinkedIn, I saw Ethan Moloch shared the paper.

我当时想，哦，Ethan，这是一篇挺差的论文。

And I was like, Oh, this is kind of a crummy paper, Ethan.

我重新分享了它，并且基本上说：这就是为什么我们不必在意这篇论文。

And I reshared it and I basically said, Here's why we don't have to care about this paper.

我详细列出了对这项研究的严厉批评。

And I outlined like a very critical review of the study.

我之所以能做到这一点，唯一的原因是我有这个系统，而且我已经看过刚发表的这篇论文了。

And the only reason why I could do that is because I had this system and I'd already looked at the paper that had just come out.

我早就已经分析过它、批评过它，并且思考过：我们能从中学到什么吗？

I had already like analyzed it and critiqued it and like, Is there something we can learn from this?

然后我写了一篇非常详细的LinkedIn帖子，这真的是我在LinkedIn上表现最好的帖子之一。

And then I wrote a really detailed LinkedIn post about it and it's honestly one of my most best performing posts on LinkedIn ever.

那不就得了。

Well, there you go.

我必须问一下，这个是怎么被触发的？

And what I have to ask is how does this get triggered?

是自动触发的吗？是通过今天这个命令触发的吗？

Is this automatically, is this triggered off that today command?

那这个是怎么回事呢？

How does that So

我确实把它集成到了我的今天命令里，但它的运作方式是，我把它做成了一个插件。

I did integrate it into my today command, but the way it works, I built this as a plugin.

它是一个公开的代码仓库。

It is available as a public repo.

我得说，它还在测试中。

I will say it's still being tested.

目前只有一个用户。

It has one user.

我丈夫将成为第二个用户。

My husband is going to be the second user.

如果你想要成为第三个用户，我们可以让它可用。

And if you want to be the third user, we can make it available.

这是一个公开的仓库，但使用时请自行承担风险，因为它仍在开发中。

It is a public repo, but use at your own caution and still in development.

有趣的是，这个项目中真正需要AI的部分只有论文摘要，但如果没有AI，我是不可能完成这个项目的。

What's funny about this is the only part of this that really requires AI is the paper summaries, but the part I would not have been able to build this without AI.

我基本上只是向Claude解释了一下：我想实现什么功能。

I basically just explained to Claude, like, here's what I want.

我想每天自动执行一次搜索。

I wanna run a daily search.

我想把待办事项列表里的内容消化掉。

I wanna digest on my to do list.

比如，我们该怎么实现这个功能？

Like, how are we gonna make this happen?

实际上，我有两个Python脚本在后台运行。

And what's happening under the hood is I have two Python scripts.

其中一个脚本每天早上搜索档案，每周日则搜索谷歌学术。

One of them every morning searches archive, and then every Sunday searches Google Scholar.

它会追踪我们已经看过的论文和新论文，并根据我自定义的关键词配置文件进行搜索。

And it's keeping track of what papers we've already seen, what papers are new, and it's searching based on a config file of my personally defined keywords.

然后每天晚上，我还有一个第二脚本，这些都是定时任务。

And then every night, I have a second script, and these are cron jobs.

它们只是在我的电脑上按计划自动运行。

They just run on my computer, like, on a schedule.

到了晚上，第二个脚本会扫描我研究目录中的源文件夹，查找任何新的PDF文件，并为我的today命令生成一个待办列表，将列表中的所有论文交给Claude Code代理生成摘要，然后将摘要添加到这个research today文件中。

And then at night, it's that second script is looking through my source directories in my research directory for any new PDFs, and it's creating a to do list for my today command to take all the papers in that list, trigger Claude Code agents to generate the summaries, and then the summaries get added to this research today file.

那你还需要手动下载这些PDF吗？

And do you still have to download those PDFs manually?

我现在还是手动下载这些PDF，虽然搜索结果里其实已经有足够信息让我可以自动下载了，但就连我亲自挑选要总结的论文，都已经像洪水一样多了。

I am downloading those PDFs manually, and could like, there is enough information in the search results that I probably could download them automatically, but I don't it's even my hand selecting what papers I want to summarize, it's already a fire hose.

所以我需要一个过滤器。

So I want a filter.

如果我们看看我的摘要，其实我不需要读完所有这些论文。

If we look at my digest, I don't need to read all of these papers.

其中有一些会明显跳出来，让我觉得：哦，这跟我做的工作真的相关。

I there's there's some of them that are gonna pop out as like, oh, that's really relevant to what I do.

我会下载那个PDF。

I'm gonna grab that PDF.

所以我每天花五到十分钟看看这个列表，下载几篇论文，然后就抛在脑后了。

So I spend like five to ten minutes a day just looking at this and downloading papers, and then I forget about it.

到了第二天，我就会收到这些摘要。

And then the very next day, I get those summaries.

太棒了。

That's awesome.

你还提到过一些档案和这些学术资源，还有没有其他类似的来源呢？

And are there any other you know, you've you've you've spoken about sort of archives and these academic sources.

你有没有考虑过为其他有用的市场信息来源也创建这样的系统？

Have you thought about creating this for other sources of market information that are maybe useful?

是的。

Yeah.

我写了一篇博客文章，叫《Claude Code》。

So I wrote a blog post like my I wrote a blog post called, Claude Code.

这是什么？

What is it?

它有什么不同？

How it's different?

为什么非技术人员应该使用它？

And why nontechnical people should use it?

是的

Yep.

在这篇博客文章中，我描述了一个场景。

And in that blog post, I gave this scenario.

我试图从一个完全的新手，一步步达到那种神奇的时刻。

I was trying to go from like total beginner to magical moment.

因此，我设计的这个神奇时刻是：在这篇博客文章里，你将学习终端的使用、了解Claude Code，到最后，Claude Code会为你指定的竞争对手生成一份非常详细的竞争分析报告，包含详细的价格对比表和功能对比表，因为我觉得这类工作正是Claude特别擅长的。

And so the magical moment I created was in that one blog post, you learn about the terminal, you learn about Claude Code, and by the end, Claude Code has generated a very detailed competitive analysis for whatever competitors you tell it with like a detailed price comparison table, a detailed feature comparison table because I think like this is the type of stuff Claude is really good at.

它能够查询信息。

It can go query things.

它能够整合信息。

It can aggregate things.

它能够生成报告。

It can create reports.

所以我开始思考，我理想中的是，能为与我相关的LinkedIn帖子生成同样的研究报告——因为我想参与讨论、评论相关内容，但每次登录LinkedIn时，我都想把眼睛戳瞎。

And so I've been starting to think about, like, what I would love is this same research report for, like, LinkedIn posts that are relevant to my because, I want to go be part of the conversation and comment on things, But when I log into LinkedIn, I kind of want to stab my eyes out.

所以我需要一个过滤器。

And so like, I need a filter.

对吧？

Right?

LinkedIn 的 API 让这件事变得非常困难。

LinkedIn's API makes that really hard.

所以我知道我本来想说，这其实是在呼吁开发一个 LinkedIn MCP，这样我们就能通过暗色模式终端访问 LinkedIn 了。

So I know I was going to say this is a call for a LinkedIn MCP here so we can just access LinkedIn through the dark mode terminal.

但在他们这么做之前，我们得想办法绕过他们的广告推送机制。

We'll have to figure out how to like push their ads through MCP before they do that.

但我确实认为，对于那些对学术研究不感兴趣的人来说，这种应用还有很多可能性。

But I do think there are a lot of applications for this that like people that aren't interested in the academic research side.

我的意思是，我现在用 Claude 来替我搜索谷歌。

I mean, Claude, I'm using Claude to Google for me.

我几乎不再去谷歌了。

I don't really go to Google anymore.

本集由Graphite赞助播出，Graphite是一个AI驱动的代码审查平台，帮助工程团队更快地交付更高质量的软件。

This episode is brought to you by Graphite, the AI powered code review platform helping engineering teams ship higher quality software faster.

随着开发者采用AI工具，代码生成速度加快，但代码审查却未能跟上步伐。

As developers adopt AI tools, code generation is accelerating, but code review hasn't caught up.

拉取请求变得更大、更杂乱，团队在等待审查上花费的时间比实际开发还多。

PRs are getting larger, noisier, and teams are spending more time blocked on review than building.

Graphite解决了这个问题。

Graphite fixes this.

Graphite将所有代码审查的核心功能整合到一个流畅的工作流中。

Graphite brings all your code review essentials into one streamlined workflow.

堆叠式差异、更简洁直观的拉取请求页面、AI驱动的审查以及自动化合并队列，全部设计用于帮助你更快地完成审查周期。

Stacked diffs, a cleaner, more intuitive PR page, AI powered reviews, and an automated merge queue, all designed to help you move through review cycles faster.

成千上万的开发者依赖Graphite来加速审查流程，从而让你专注于开发，而不是等待。

Thousands of developers rely on Graphite to move through review faster so you can focus on building, not waiting.

立即访问 graphitedev.link/howiai 开始使用。

Check it out at graphitedev.link/howiai to get started.