pertrai1

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.

pertrai1

Oh my zsh.

Install ZSH.

sudo apt install zsh-autosuggestions zsh-syntax-highlighting zsh

Install Oh my ZSH.

pertrai1

--- Actions ---
$Copy                                              <M-C>
$Cut                                               <M-X> <S-Del>
$Delete                                            <Del> <BS> <M-BS>
$LRU                                              
$Paste                                             <M-V>
$Redo                                              <M-S-Z> <A-S-BS>
$SearchWeb                                         <A-S-G>
$SelectAll                                         <M-A>
$Undo                                              <M-Z>

pertrai1

{
  "final_space": true,
  "console_title": true,
  "console_title_style": "folder",
  "blocks": [
    {
      "type": "prompt",
      "alignment": "left",
      "horizontal_offset": 0,
      "vertical_offset": 0,

pertrai1

const response = await fetch(`https://catappapi.herokuapp.com/users/${userId}`)
  const data = await response.json()
  return data.imageUrl
}

pertrai1

import { run } from 'runjs'

const task = {
  'clean': () => {
    run('rm -rf node_modules')
    run('rm -rf build')
  },
  'build:clean': () => {
    run('rm -rf build')
    run('mkdir build')

pertrai1

"scripts": {
    "clean": "rimraf dist/*",
    "prebuild": "npm run clean -s",
    "build": "npm run build:scripts -s && npm run build:styles -s && npm run build:markup -s",
    "build:scripts": "browserify -d assets/scripts/main.js -p [minifyify --compressPath . --map main.js.map --output dist/main.js.map] | hashmark -n dist/main.js -s -l 8 -m assets.json 'dist/{name}{hash}{ext}'",
    "build:styles": "stylus assets/styles/main.styl -m -o dist/ && hashmark -s -l 8 -m assets.json dist/main.css 'dist/{name}{hash}{ext}'",
    "build:markup": "jade assets/markup/index.jade --obj assets.json -o dist",
    "test": "karma start --singleRun",
    "watch": "parallelshell 'npm run watch:test -s' 'npm run watch:build -s'",
    "watch:test": "karma start",

pertrai1

tmux cheat sheet

(C-x means ctrl+x, M-x means alt+x)

Prefix key

The default prefix is C-b. If you (or your muscle memory) prefer C-a, you need to add this to ~/.tmux.conf:

remap prefix to Control + a

pertrai1

#!/usr/bin/env ruby
#
# This script is an astonishing feat of top notch
# rockstar craftsmanship. It totally uses artificial
# intelligence to extract colors out of tmTheme and
# build an itermcolors scheme file for iTerm2.
#
# I know this sounds crazy, but it actually knows
# approximately what colors should be used in the
# ANSI list, and tries to find nearest colors from

pertrai1

Quick Git Reference

Configuring

git config --global user.name "<First Last>"
git config --global user.email <johndoe@example.com>
git config --global core.editor <vim>

git config --global color.ui true