章节 1：引言——基于AI的个人知识库系统

📝 本节摘要：

本节介绍了利用AI构建个人知识库的核心理念：抛弃分散的笔记软件，通过单一文件夹结合AI实现自动整理、总结和互联。作者预览了搭建该系统所需的核心要素与时间成本，并提供了一键构建该系统的工具链接。

[原文] [Author]: @karpathy dropped a post describing how he uses AI to build personal knowledge bases.

[译文] [Author]: @karpathy 发布了一篇帖子，描述了他如何使用人工智能（AI）来构建个人知识库。

[原文] [Author]: The idea is simple: instead of keeping notes scattered across apps, you dump everything into one folder.

[译文] [Author]: 这个想法很简单：与其将笔记散落在各个应用程序（apps）中，不如将所有内容都扔进一个文件夹中。

[原文] [Author]: Then you tell your AI to organize all of it into a personal wiki - summaries, connections, articles - that gets smarter every time you use it.

[译文] [Author]: 然后，你告诉你的 AI 将所有这些组织成一个个人维基（wiki）——包含摘要、关联和文章——它会在你每次使用时变得更聪明。

[原文] [Author]: No special software.

[译文] [Author]: 不需要特殊的软件。

[原文] [Author]: No database.

[译文] [Author]: 不需要数据库。

[原文] [Author]: Just folders and text files.

[译文] [Author]: 只需要文件夹和文本文件。

[原文] [Author]: In under 7 minutes you'll learn:

[译文] [Author]: 在不到 7 分钟的时间里，你将学到：

[原文] [Author]: The exact folder structure to set up (takes 2 minutes)

[译文] [Author]: 需要设置的具体文件夹结构（耗时 2 分钟）

[原文] [Author]: How to automate web scraping into your knowledge base with one CLI tool

[译文] [Author]: 如何使用一个命令行界面（CLI）工具将网页抓取（web scraping）自动化到你的知识库中

[原文] [Author]: The one-file "schema" that makes the whole system work

[译文] [Author]: 使整个系统运转的单文件“模式”（schema）

[原文] [Author]: How to get your AI to compile raw notes into an organized wiki

[译文] [Author]: 如何让你的 AI 将原始笔记编译成一个有条理的维基（wiki）

[原文] [Author]: The compounding trick that makes it smarter every time you use it

[译文] [Author]: 每次使用时都能让它变得更聪明的复利技巧

[原文] [Author]: The health check that catches mistakes before they pile up

[译文] [Author]: 在错误堆积之前捕捉错误的健康检查（health check）

[原文] [Author]: Follow @NickSpisak_ for practitioner-level AI implementation and Claude Code architecture content.

[译文] [Author]: 关注 @NickSpisak_ 以获取从业者级别的 AI 实施和 Claude Code 架构内容。

[原文] [Author]: Want the skill that builds this system for you?

[译文] [Author]: 想要能为你构建这个系统的技能吗？

[原文] [Author]: Grab the free LLM Knowledge Base skill here: https://return-my-time.kit.com/286e11f7e6

[译文] [Author]: 在这里获取免费的大语言模型知识库（LLM Knowledge Base）技能：https://return-my-time.kit.com/286e11f7e6

[原文] [Author]: Alright let's get started...

[译文] [Author]: 好了，让我们开始吧……

章节 2：步骤 1 与 2——建立目录与收集原始素材

📝 本节摘要：

本节详细说明了系统搭建的第一步和第二步。首先只需创建一个包含 raw、wiki 和 outputs 的三层子目录结构；接着，用户需将所有未整理的素材（笔记、截图、论文等）全部倾倒至 raw 文件夹中，强调这部分无需人为进行任何整理工作。

[原文] [Author]: 1. Create Three Folders (2 Minutes)

[译文] [Author]: 1. 创建三个文件夹（耗时 2 分钟）

[原文] [Author]: Open your terminal or file explorer.

[译文] [Author]: 打开你的终端（terminal）或文件资源管理器（file explorer）。

[原文] [Author]: Create a project folder anywhere on your computer.

[译文] [Author]: 在你的电脑上的任何位置创建一个项目文件夹。

[原文] [Author]: Inside it, create three subfolders:

[译文] [Author]: 在里面创建三个子文件夹：

[原文] [Author]: my-knowledge-base/

[译文] [Author]: my-knowledge-base/（我的知识库/）

[原文] [Author]: raw/ (your source material - articles, notes, screenshots)

[译文] [Author]: raw/ （你的原始素材 - 文章、笔记、截图）

[原文] [Author]: wiki/ (where your AI will write the organized version)

[译文] [Author]: wiki/ （你的 AI 编写整理后版本的地方）

[原文] [Author]: outputs/ (answers, reports, and research your AI generates)

[译文] [Author]: outputs/ （你的 AI 生成的答案、报告和研究的地方）

[原文] [Author]: That's it.

[译文] [Author]: 就是这样。

[原文] [Author]: This is the same structure @karpathy uses.

[译文] [Author]: 这是 @karpathy 使用的相同结构。

[原文] [Author]: The raw/ folder is your junk drawer of source material.

[译文] [Author]: raw/ 文件夹是你存放原始素材的“杂物屉”。

[原文] [Author]: The wiki/ folder is where the AI turns that mess into something organized.

[译文] [Author]: wiki/ 文件夹是 AI 将这些杂乱内容转化为有条理内容的地方。

[原文] [Author]: The outputs/ folder is where answers to your questions live.

[译文] [Author]: outputs/ 文件夹是存放你问题答案的地方。

[原文] [Author]: No apps to install.

[译文] [Author]: 没有需要安装的应用程序。

[原文] [Author]: No accounts to create.

[译文] [Author]: 没有需要创建的账户。

[原文] [Author]: Three folders.

[译文] [Author]: 只有三个文件夹。

[原文] [Author]: 2. Fill Your Raw Folder (10 Minutes)

[译文] [Author]: 2. 填充你的 Raw（原始）文件夹（耗时 10 分钟）

[原文] [Author]: This is where most people stall.

[译文] [Author]: 这是大多数人停滞不前的地方。

[原文] [Author]: They create the folders and then stare at an empty raw/ directory wondering what to put in it.

[译文] [Author]: 他们创建了文件夹，然后盯着一个空的 raw/ 目录，不知道该放些什么进去。

[原文] [Author]: The answer: everything.

[译文] [Author]: 答案是：所有东西。

[原文] [Author]: Copy-paste articles into .md or .txt files.

[译文] [Author]: 将文章复制粘贴到 .md 或 .txt 文件中。

[原文] [Author]: Save screenshots or diagrams as images.

[译文] [Author]: 将截图或图表保存为图像。

[原文] [Author]: Export notes from whatever app you use now.

[译文] [Author]: 从你现在使用的任何应用中导出笔记。

[原文] [Author]: Paste in meeting notes, research papers, project docs.

[译文] [Author]: 粘贴会议笔记、研究论文、项目文档。

[原文] [Author]: Dump in bookmarks you've been hoarding for months.

[译文] [Author]: 倒入你囤积了几个月的书签。

[原文] [Author]: Don't organize it.

[译文] [Author]: 不要去整理它。

[原文] [Author]: Don't rename anything.

[译文] [Author]: 不要重命名任何东西。

[原文] [Author]: Don't clean it up.

[译文] [Author]: 不要清理它。

[原文] [Author]: That's the AI's job.

[译文] [Author]: 那是 AI 的工作。

[原文] [Author]: I keep 17 raw source files across my X content pipeline - clipped articles, competitor breakdowns, analytics reports.

[译文] [Author]: 我在我的 X（推特）内容管道中保留了 17 个原始源文件——剪报文章、竞争对手分析、数据分析报告。

[原文] [Author]: None of it is organized by hand.

[译文] [Author]: 没有一样是手工整理的。

[原文] [Author]: But Karpathy didn't mention the part that actually speeds this up: automated collection.

[译文] [Author]: 但是 Karpathy 没有提到真正能加速这个过程的部分：自动化收集。

章节 3：步骤 3——自动化收集与 Agent Browser 工具

📝 本节摘要：

本节介绍了如何使用 Vercel Labs 的 agent-browser 命令行工具来实现素材的自动化收集。该工具允许 AI 直接控制浏览器抓取动态页面或复杂内容并一键保存，极大提升了填充 raw 文件夹的效率。

[原文] [Author]: 3. Automate Source Collection With Agent Browser (Optional But Powerful)

[译文] [Author]: 3. 使用 Agent Browser 自动化来源收集（可选但强大）

[原文] [Author]: Vercel Labs just shipped agent-browser - a free CLI tool that lets your AI agent control a real browser.

[译文] [Author]: Vercel Labs 刚刚发布了 agent-browser ——一个允许你的 AI 代理（AI agent）控制真实浏览器的免费命令行（CLI）工具。

[原文] [Author]: 26K+ GitHub stars.

[译文] [Author]: 在 GitHub 上有超过 2.6 万颗星。

[原文] [Author]: Two commands to install:

[译文] [Author]: 两个安装命令：

[原文] [Author]: npm install -g agent-browser

[译文] [Author]: npm install -g agent-browser

[原文] [Author]: agent-browser install

[译文] [Author]: agent-browser install

[原文] [Author]: That second command downloads a dedicated Chrome browser.

[译文] [Author]: 第二个命令会下载一个专用的 Chrome 浏览器。

[原文] [Author]: Now your AI can scrape any webpage, extract the text, and save it directly into your raw/ folder.

[译文] [Author]: 现在你的 AI 可以抓取任何网页，提取文本，并将其直接保存到你的 raw/ 文件夹中。

[原文] [Author]: Here's what that looks like in practice:

[译文] [Author]: 实际操作如下：

[原文] [Author]: agent-browser open https://some-article-you-want.com

[译文] [Author]: agent-browser open https://some-article-you-want.com

[原文] [Author]: agent-browser get text "article"

[译文] [Author]: agent-browser get text "article"

[原文] [Author]: That's it.

[译文] [Author]: 就是这样。

[原文] [Author]: The AI opens the page, grabs the article text, and you pipe it into a file in raw/.

[译文] [Author]: AI 打开页面，抓取文章文本，然后你将其导入到 raw/ 中的一个文件里。

[原文] [Author]: No manual copy-paste.

[译文] [Author]: 不需要手动复制粘贴。

[原文] [Author]: No browser extensions.

[译文] [Author]: 不需要浏览器扩展程序。

[原文] [Author]: Agent-browser handles pages that copy-paste can't touch.

[译文] [Author]: Agent-browser 可以处理复制粘贴无法触及的页面。

[原文] [Author]: JavaScript-heavy sites that load content dynamically.

[译文] [Author]: 动态加载内容的重度 JavaScript 网站。

[原文] [Author]: Pages behind logins.

[译文] [Author]: 需要登录才能访问的页面。

[原文] [Author]: Research papers with interactive figures.

[译文] [Author]: 带有交互式图表的研究论文。

[原文] [Author]: Anything that requires scrolling, clicking "load more," or navigating through menus before the content appears.

[译文] [Author]: 任何在内容出现之前需要滚动、点击“加载更多”或在菜单中导航的操作。

[原文] [Author]: The tool uses 82% fewer tokens than Playwright MCP, which means your AI agent can scrape 5-6x more pages within the same session.

[译文] [Author]: 该工具使用的代币（tokens）比 Playwright MCP 少 82%，这意味着你的 AI 代理在同一会话中可以多抓取 5-6 倍的页面。

[原文] [Author]: I use it to pull competitor articles, trending threads, and research docs directly into my pipeline without ever opening a browser myself.

[译文] [Author]: 我用它将竞争对手的文章、热门帖子和研究文档直接拉入我的管道中，而我自己根本不需要打开浏览器。

[原文] [Author]: For your knowledge base, the workflow is simple: find an article you want, tell your AI "scrape this URL and save it to raw/", and agent-browser handles the rest.

[译文] [Author]: 对于你的知识库来说，工作流程很简单：找到你想要的文章，告诉你的 AI “抓取这个 URL 并将其保存到 raw/”，agent-browser 会处理剩下的事情。

[原文] [Author]: Your raw/ folder fills itself.

[译文] [Author]: 你的 raw/ 文件夹会自动填满。

章节 4：步骤 4——编写 Schema 规则文件

📝 本节摘要：

本节指出为系统编写一个 Schema 配置文件（如 CLAUDE.md）是至关重要的步骤。文中直接提供了一份开箱即用的模板，指导 AI 如何严格管理各级文件夹、生成带摘要的特定文件结构并实现知识维基的索引和互联。

[原文] [Author]: 4. Write Your Schema File (5 Minutes)

[译文] [Author]: 4. 编写你的 Schema 文件（耗时 5 分钟）

[原文] [Author]: This is the part most people will skip. Don't.

[译文] [Author]: 这是大多数人会跳过的部分。别跳过。

[原文] [Author]: Create a file in the root of your project called CLAUDE.md (or AGENTS.md or README.md - the name doesn't matter, the content does).

[译文] [Author]: 在你的项目根目录下创建一个名为 CLAUDE.md（或 AGENTS.md 或 README.md——名字不重要，内容才重要）的文件。

[原文] [Author]: This file tells your AI what the knowledge base is about and how to organize it.

[译文] [Author]: 这个文件告诉你的 AI 知识库是关于什么的，以及如何组织它。

[原文] [Author]: Here's a starter template you can copy right now:

[译文] [Author]: 这是一个你现在就可以直接复制的入门模板：

[原文] [Author]: markdown

[译文] [Author]: markdown

[原文] [Author]: ### Knowledge Base Schema

[译文] [Author]: ### 知识库模式（Schema）

[原文] [Author]: #### What This Is

[译文] [Author]: #### 这是什么

[原文] [Author]: A personal knowledge base about [YOUR TOPIC].

[译文] [Author]: 一个关于 [你的主题] 的个人知识库。

[原文] [Author]: #### How It's Organized

[译文] [Author]: #### 它是如何组织的

[原文] [Author]: * raw/ contains unprocessed source material. Never modify these files.

[译文] [Author]: * raw/ 包含未处理的原始素材。永远不要修改这些文件。

[原文] [Author]: * wiki/ contains the organized wiki. AI maintains this entirely.

[译文] [Author]: * wiki/ 包含整理好的维基（wiki）。这完全由 AI 维护。

[原文] [Author]: * outputs/ contains generated reports, answers, and analyses.

[译文] [Author]: * outputs/ 包含生成的报告、答案和分析。

[原文] [Author]: #### Wiki Rules

[译文] [Author]: #### 维基规则（Wiki Rules）

[原文] [Author]: * Every topic gets its own .md file in wiki/

[译文] [Author]: * 每个主题在 wiki/ 中都有自己的 .md 文件

[原文] [Author]: * Every wiki file starts with a one-paragraph summary

[译文] [Author]: * 每个维基文件都以一段摘要开始

[原文] [Author]: * Link related topics to each other using [[topic-name]] format

[译文] [Author]: * 使用 [[topic-name]] 格式将相关主题相互链接

[原文] [Author]: * Maintain an INDEX.md in wiki/ that lists every topic with a one-line description

[译文] [Author]: * 在 wiki/ 中维护一个 INDEX.md，列出每个主题并附带一行描述

[原文] [Author]: * When new raw sources are added, update the relevant wiki articles

[译文] [Author]: * 当添加新的原始素材时，更新相关的维基文章

[原文] [Author]: #### My Interests

[译文] [Author]: #### 我的兴趣

[原文] [Author]: [List 3-5 things you want this knowledge base to focus on]

[译文] [Author]: [列出 3-5 件你希望这个知识库重点关注的事情]

[原文] [Author]: @karpathy confirmed he keeps his schema "super simple and flat" in an AGENTS.md file.

[译文] [Author]: @karpathy 证实他将他的模式（schema）保存在一个 AGENTS.md 文件中，并保持其“极其简单且扁平”。

[原文] [Author]: No database.

[译文] [Author]: 没有数据库。

[原文] [Author]: No plugin.

[译文] [Author]: 没有插件。

[原文] [Author]: Just a text file that tells the AI the rules.

[译文] [Author]: 只是一个告诉 AI 规则的文本文件。

[原文] [Author]: This is the equivalent of what I use CLAUDE.md for across every project.

[译文] [Author]: 这相当于我在每个项目中使用的 CLAUDE.md 的作用。

[原文] [Author]: It's the AI's instruction manual for your specific knowledge base.

[译文] [Author]: 它是针对你特定知识库的 AI 指导手册。

章节 5：步骤 5 与 6——AI 编译维基与交互式提问

📝 本节摘要：

这一段介绍了核心的执行过程：通过向 AI 开发工具下达编译指令，让 AI 根据 Schema 自动完成知识库的建立和链接。构建完成后，用户就可以通过提问与系统互动，且每次的回答和见解都可以再度存入库中，实现知识的闭环和复利。

[原文] [Author]: 5. Tell Your AI to Compile the Wiki (15 Minutes)

[译文] [Author]: 5. 告诉你的 AI 编译维基（耗时 15 分钟）

[原文] [Author]: Open Claude Code (or Cursor, or any AI coding tool that can read your files).

[译文] [Author]: 打开 Claude Code（或 Cursor，或任何可以读取你文件的 AI 编码工具）。

[原文] [Author]: Point it at your project folder and say:

[译文] [Author]: 将其指向你的项目文件夹并说：

[原文] [Author]: "Read everything in raw/. Then compile a wiki in wiki/ following the rules in CLAUDE.md. Create an INDEX.md first, then create one .md file per major topic. Link related topics. Summarize every source."

[译文] [Author]: “读取 raw/ 中的所有内容。然后遵循 CLAUDE.md 中的规则在 wiki/ 中编译一个维基。首先创建一个 INDEX.md，然后为每个主要主题创建一个 .md 文件。链接相关主题。总结每个来源。”

[原文] [Author]: Then step away. Let it work.

[译文] [Author]: 然后离开。让它工作。

[原文] [Author]: When it's done, you'll have a wiki/ folder full of organized articles - connections you didn't see, summaries of things you forgot you saved, and an index file that makes everything searchable in seconds.

[译文] [Author]: 完成后，你将拥有一个满是整理好文章的 wiki/ 文件夹——包含了你没有注意到的关联、你忘记曾保存过内​​容的摘要，以及一个让所有内容能在几秒钟内被搜索到的索引文件。

[原文] [Author]: The important thing: you don't edit the wiki by hand.

[译文] [Author]: 重要的一点：你不要手动编辑维基。

[原文] [Author]: That's the AI's job.

[译文] [Author]: 那是 AI 的工作。

[原文] [Author]: You read it, you ask questions against it, and the AI keeps it updated.

[译文] [Author]: 你阅读它，你针对它提问，然后 AI 会保持它的更新。

[原文] [Author]: 6. Ask Questions and Save the Answers (Ongoing)

[译文] [Author]: 6. 提问并保存答案（持续进行）

[原文] [Author]: Once your wiki has 10+ articles, start asking questions:

[译文] [Author]: 一旦你的维基有 10 篇以上的文章，就可以开始提问了：

[原文] [Author]: "Based on everything in wiki/, what are the three biggest gaps in my understanding of [topic]?"

[译文] [Author]: “基于 wiki/ 中的所有内容，我在对 [主题] 的理解上最大的三个空白是什么？”

[原文] [Author]: "Compare what source A says about [concept] vs what source B says. Where do they disagree?"

[译文] [Author]: “比较来源 A 和来源 B 对 [概念] 的说法。它们在哪些地方存在分歧？”

[原文] [Author]: "Write me a 500-word briefing on [topic] using only what's in this knowledge base."

[译文] [Author]: “只使用这个知识库中的内容，为我写一份关于 [主题] 的 500 字简报。”

[原文] [Author]: The AI reads across your entire wiki and gives you answers grounded in your own collected material.

[译文] [Author]: AI 会通读你的整个维基，并根据你自己收集的素材为你提供答案。

[原文] [Author]: Save those answers back into the knowledge base.

[译文] [Author]: 将这些答案保存回知识库中。

[原文] [Author]: Drop the output into outputs/ or have the AI update the relevant wiki article with the new insight.

[译文] [Author]: 将输出放到 outputs/ 中，或者让 AI 用新的见解更新相关的维基文章。

[原文] [Author]: Every question makes the next answer better.

[译文] [Author]: 每一个问题都会让下一个答案变得更好。

[原文] [Author]: That's the loop.

[译文] [Author]: 这就是一个循环。

章节 6：步骤 7、8 与结语——定期体检与工具极简主义

📝 本节摘要：

最后的部分讨论了系统的维护与工具理念。为了防止 AI 产生的错误在库中形成“负面复利”，建议每月让 AI 运行一次矛盾与逻辑空白的自查。同时，作者强调工具极简主义，指出简单的扁平文件结构远胜于依赖几十个插件的繁重工具，呼吁大家停止寻找完美工具，立刻动手搭建。

[原文] [Author]: 7. Run a Health Check (Monthly)

[译文] [Author]: 7. 运行健康检查（每月一次）

[原文] [Author]: Tell your AI:

[译文] [Author]: 告诉你的 AI：

[原文] [Author]: "Review the entire wiki/ directory. Flag any contradictions between articles. Find topics mentioned but never explained. List any claims that aren't backed by a source in raw/. Suggest 3 new articles that would fill gaps."

[译文] [Author]: “查看整个 wiki/ 目录。标记文章之间的任何矛盾。找出被提及但从未解释过的主题。列出 raw/ 中没有原始来源支持的任何主张。建议 3 篇可以填补空白的新文章。”

[原文] [Author]: One of the best replies to Karpathy's post came from @HFloyd : "When outputs get filed back, errors compound too."

[译文] [Author]: 对 Karpathy 帖子的最好回复之一来自 @HFloyd ：“当输出被归档回去时，错误也会复利增加。”

[原文] [Author]: That's real.

[译文] [Author]: 这是非常真实的。

[原文] [Author]: If the AI writes something slightly wrong and you save it back, the next answer builds on that wrong thing.

[译文] [Author]: 如果 AI 写了稍微错误的内容而你将其保存回去，下一个答案就会建立在这个错误的内容之上。

[原文] [Author]: The fix is simple: run periodic health checks.

[译文] [Author]: 解决方法很简单：运行定期的健康检查。

[原文] [Author]: 8. You Don't Need Obsidian (But You Can Use It)

[译文] [Author]: 8. 你不需要 Obsidian（但你可以使用它）

[原文] [Author]: Half the replies to Karpathy's post were people pitching Obsidian plugins.

[译文] [Author]: Karpathy 帖子下一半的回复都是人们在推销 Obsidian 插件。

[原文] [Author]: Lex Fridman uses Obsidian + Cursor.

[译文] [Author]: Lex Fridman 使用 Obsidian + Cursor。

[原文] [Author]: There are already three startups building dedicated tools for this.

[译文] [Author]: 已经有三家初创公司在为此构建专用工具。

[原文] [Author]: But here's what Karpathy actually said when someone asked about his setup:

[译文] [Author]: 但当有人问及他的设置时，Karpathy 实际上是这样说的：

[原文] [Author]: "I'm trying to keep it super simple and flat. It's just a nested directory of .md files."

[译文] [Author]: “我试图保持它极其简单且扁平。它只是一个包含 .md 文件的嵌套目录。”

[原文] [Author]: A folder of text files and a schema file is the entire product.

[译文] [Author]: 一个包含文本文件的文件夹和一个模式（schema）文件就是整个产品。

[原文] [Author]: I run my whole knowledge system from the terminal with Claude Code.

[译文] [Author]: 我使用 Claude Code 从终端运行我的整个知识系统。

[原文] [Author]: You could use VS Code.

[译文] [Author]: 你可以使用 VS Code。

[原文] [Author]: You could use Obsidian.

[译文] [Author]: 你可以使用 Obsidian。

[原文] [Author]: You could use Notepad.

[译文] [Author]: 你可以使用记事本。

[原文] [Author]: The AI doesn't care what app you open the files in.

[译文] [Author]: AI 并不关心你用什么应用打开文件。

[原文] [Author]: What matters is the folder structure and the schema.

[译文] [Author]: 重要的是文件夹结构和模式（schema）。

[原文] [Author]: Obsidian with 47 plugins is the Notion trap all over again.

[译文] [Author]: 安装了 47 个插件的 Obsidian 又是一次 Notion 陷阱的重演。

[原文] [Author]: You spend more time configuring the tool than using the knowledge base.

[译文] [Author]: 你花在配置工具上的时间比使用知识库的时间还要多。

[原文] [Author]: Flat files and a good schema will outperform a fancy tool stack 90% of the time.

[译文] [Author]: 扁平化的文件和良好的模式在 90% 的情况下都会胜过花哨的工具栈。

[原文] [Author]: I've watched it happen across dozens of client setups.

[译文] [Author]: 我在几十个客户的系统设置中见证了这种情况的发生。

[原文] [Author]: Stop shopping for the perfect tool. Start building.

[译文] [Author]: 停止寻找完美的工具吧。开始构建。

[原文] [Author]: That's the full system.

[译文] [Author]: 这就是完整的系统。

[原文] [Author]: Three folders, one schema file, a browser scraper, and an AI that maintains everything.

[译文] [Author]: 三个文件夹，一个模式文件，一个浏览器抓取工具，以及一个维护一切的 AI。

[原文] [Author]: 41K people bookmarked Karpathy's post.

[译文] [Author]: 有 4.1 万人收藏了 Karpathy 的帖子。

[原文] [Author]: The difference between bookmarking it and benefiting from it is one weekend of setup.

[译文] [Author]: 收藏它和从中受益之间的区别，只在于一个周末的设置时间。

[原文] [Author]: Pick your topic.

[译文] [Author]: 挑选你的主题。

[原文] [Author]: Create the folders.

[译文] [Author]: 创建文件夹。

[原文] [Author]: Drop in what you already have.

[译文] [Author]: 投入你已有的素材。

[原文] [Author]: Let the AI do the rest.

[译文] [Author]: 让 AI 处理剩下的事情。

[原文] [Author]: Want to skip the manual setup?

[译文] [Author]: 想跳过手动设置吗？

[原文] [Author]: Grab the free LLM Knowledge Base skill that scaffolds the entire system in 60 seconds: https://return-my-time.kit.com/286e11f7e6

[译文] [Author]: 获取在 60 秒内搭建整个系统框架的免费 LLM 知识库技能：https://return-my-time.kit.com/286e11f7e6