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.

jake-stewart

Terminals should generate the 256-color palette from the user's base16 theme.

If you've spent much time in the terminal, you've probably set a custom base16 theme. They work well. You define a handful of colors in one place and all your programs use them.

The drawback is that 16 colors is limiting. Complex and color-heavy programs struggle with such a small palette.

jasonsnell

import subprocess
import re
import sys
from Levenshtein import ratio
from collections import defaultdict, Counter

# Configuration: Adjust as needed
INCLUDE_PARENTHESES = False  # Set to True to include parentheticals, False to exclude them
OUTPUT_TO_CLIPBOARD = True  # Set to True to copy output to clipboard, False to print to console

ttscoff

#!/usr/bin/env ruby
# Archive linkding bookmarks to Markdown files
# Can use [Gather](https://brettterpstra.com/projects/gather-cli/)
# for conversion (if installed), or use Marky
# the Markdownifier (web-based).
#
# See options below for configuration
#
# This script is designed to run once initially, and then
# be set to run in the background at intervals using cron

OrionReed

// 3D Dom viewer, copy-paste this into your console to visualise the DOM as a stack of solid blocks.
// You can also minify and save it as a bookmarklet (https://www.freecodecamp.org/news/what-are-bookmarklets/)
(() => {
  const SHOW_SIDES = false; // color sides of DOM nodes?
  const COLOR_SURFACE = true; // color tops of DOM nodes?
  const COLOR_RANDOM = false; // randomise color?
  const COLOR_HUE = 190; // hue in HSL (https://hslpicker.com)
  const MAX_ROTATION = 180; // set to 360 to rotate all the way round
  const THICKNESS = 20; // thickness of layers
  const DISTANCE = 10000; // ¯\\_(ツ)_/¯

chockenberry

#!/bin/sh

if [ ! -z "$*" ]; then
	echo "this is ~/bin/df, use /bin/df"
	exit 1
fi

protect=`mount | grep -v "read-only" | grep "protect" | cut -f 3 -w`
nosuid=`mount | grep -v "read-only" | grep "nosuid" | cut -f 3 -w`

theacodes

import threading

ANIMALS = list(
    "🐶🐱🐭🐹🐰🦊🐻🐼🐨🐯🦁🐮🐷🐽🐸🐵🐔🐧🐦🐤🪿🦆🦉🦇🦇🐺🐗🐴🦄🫎🐝🪱🐛🦋🐌🐞🐜🪰🪲🪳🦟🦗🕷️🕸️🦂🐢🐍🦎🦖🦕🐙🦑🪼🦐🦞🦀🐡🐠🐟🐬🐳🐋🦈🦭🐊🐅🐆🦓🦍🦧🦣🐘🦛🦏🐪🐫🦒🦘🦬🐃🐂🐄🫏🐎🐖🐏🐑🦙🐐🦌🐕🐩🐈🐓🦃🦤🦚🦜🦢🦩🕊️🐇🦝🦨🦡🦫🦦🦥🐁🐀🐿️🦔🐉🐲"
)
ANIMALS_LEFT = []
THREAD_EMOJI = {}


def thread_emoji():

melroy89

Mastodon Docker Setup

In this guide we will only focus on using the prebuilt images from Docker Hub.

Prerequisites: You have Git, Docker, Docker compose and Nginx pre-installed.

Prepare

Clone Mastodon's repository.

jasonsnell

#! /usr/bin/env python3

import time
import jwt
import json
import requests
from datetime import datetime
from collections import defaultdict
import matplotlib.pyplot as plt

aleksbrgt

/*
This is the c configuration file for the keymap
Copyright 2012 Jun Wako <wakojun@gmail.com>
Copyright 2015 Jack Humbert
This program is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 2 of the License, or
(at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of