karpathy

LLM Wiki

A pattern for building personal knowledge bases using LLMs.

This is an idea file, it is designed to be copy pasted to your own LLM Agent (e.g. OpenAI Codex, Claude Code, OpenCode / Pi, or etc.). Its goal is to communicate the high level idea, but your agent will build out the specifics in collaboration with you.

The core idea

Most people's experience with LLMs and documents looks like RAG: you upload a collection of files, the LLM retrieves relevant chunks at query time, and generates an answer. This works, but the LLM is rediscovering knowledge from scratch on every question. There's no accumulation. Ask a subtle question that requires synthesizing five documents, and the LLM has to find and piece together the relevant fragments every time. Nothing is built up. NotebookLM, ChatGPT file uploads, and most RAG systems work this way.

zhangdxchn

> find / -name FILE_NAME
> ps -ef | grep tomcat | awk '{print "kill -9 " $2}' | sh
> ln -s full-path-to-mysql-os  /usr/local/mysql
> unlink /usr/local/mysql
> curl -XGET --header 'Content-Type: application/json'
> curl -XGET -H 'Content-Type: application/json'

# 打印目录树中文
> tree -N

paulirish

Option 1: Command-line download extension as zip and extract

extension_id=jifpbeccnghkjeaalbbjmodiffmgedin   # change this ID
curl -L -o "$extension_id.zip" "https://clients2.google.com/service/update2/crx?response=redirect&os=mac&arch=x86-64&nacl_arch=x86-64&prod=chromecrx&prodchannel=stable&prodversion=44.0.2403.130&x=id%3D$extension_id%26uc" 
unzip -d "$extension_id-source" "$extension_id.zip"

Thx to crxviewer for the magic download URL.

GiaoGiaoCat

// 检测浏览器的 User Agent 应该是非常简单的事情
// 微信在 Android 下的 User Agent
mozilla/5.0 (linux; u; android 4.1.2; zh-cn; mi-one plus build/jzo54k) applewebkit/534.30 (khtml, like gecko) version/4.0 mobile safari/534.30 micromessenger/5.0.1.352

// 微信在 iPhone 下的 User Agent
mozilla/5.0 (iphone; cpu iphone os 5_1_1 like mac os x) applewebkit/534.46 (khtml, like gecko) mobile/9b206 micromessenger/5.0

// 通过javascript判断
// 很容易看出来，微信的 User Agent 都有‘micromessenger’字符串标示，我们判断是否含有这些字符串就OK了
function isWeixinBrowser(){

gmccreight

" copy all this into a vim buffer, save it, then...
" source the file by typing :so %
" Now the vim buffer acts like a specialized application for mastering vim

" There are two queues, Study and Known.  Depending how confident you feel
" about the item you are currently learning, you can move it down several
" positions, all the way to the end of the Study queue, or to the Known
" queue.

" type ,, (that's comma comma)

theconektd

body {
  font-family: Helvetica, arial, sans-serif;
  font-size: 14px;
  line-height: 1.6;
  padding-top: 10px;
  padding-bottom: 10px;
  background-color: white;
  padding: 30px; }

body > *:first-child {