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.

brizandrew

#!/usr/bin/env python
# -*- coding: utf-8 -*-

from flask import Flask, request
from urllib import unquote_plus
import json
import re


app = Flask(__name__)

allenyllee

// import_json_appsscript.js
// https://gist.github.com/allenyllee/c764c86ed722417948fc256b7a5077c4
//
// Changelog:
// (Oct. 16 2019) tag: allenyllee-20191016
// 1. Fixed google script error: urlfetchapp - service invoked too many times https://stackoverflow.com/questions/10598179/google-apps-script-urlfetchapp-service-invoked-too-many-times
// (Jul. 16 2018) tag: allenyllee-20180716
// 1. Fixed the issue "If you try to query /arrayA[k]/arrayB[n]/arrayC[m]/.../member, you will always get /arrayA[k]/arrayB[k]/arrayC[k]/.../member."
// (Nov. 30 2017) tag: allenyllee-20171130
// 1. Add the ability to query array elements by using xpath like "/array[n]/member" where "n" is array index

jashkenas

var active = false;

function changeRefer(details) {
  if (!active) return;
  
  for (var i = 0; i < details.requestHeaders.length; ++i) {
    if (details.requestHeaders[i].name === 'Referer') {
      details.requestHeaders[i].value = 'http://www.google.com/';
      break;
    }

paulgambill

/**
 * Retrieves all the rows in the active spreadsheet that contain data and logs the
 * values for each row.
 * For more information on using the Spreadsheet API, see
 * https://developers.google.com/apps-script/service_spreadsheet
 */
function readRows() {
  var sheet = SpreadsheetApp.getActiveSheet();
  var rows = sheet.getDataRange();
  var numRows = rows.getNumRows();

vs4vijay

function getDataFromXpath(path, url) {
  var data = UrlFetchApp.fetch(url);
  var text = data.getContentText();
  var xmlDoc = Xml.parse(text, true);
  
  // Replacing tbody tag because app script doesnt understand.
  path = path.replace("/html/","").replace("/tbody","","g");
  var tags = path.split("/");
  Logger.log("tags : " + tags);
  // getting the DOM of HTML

johnjreiser

import urllib, os, sys, zipfile

def download(url,name=""):
    if(name == ""):
        name = url.split('/')[-1]
    webFile = urllib.urlopen(url)
    localFile = open(name, 'w')
    fname = localFile.name
    localFile.write(webFile.read())
    webFile.close()

ndarville

Business Models

Advertising

Models	Examples
Display ads	Yahoo!
Search ads	Google

jhaus

function gup( name ){
var cre_getDocIdelement = document.getElementById( 'gbmpm_0_l' );

name = name.replace( /[\[]/,"\\\[" ).replace( /[\]]/,"\\\]" );
    var regexS  = "[\\?&]" + name + "=([^&#]*)";
    var regex   = new RegExp( regexS );
    var results = regex.exec( cre_getDocIdelement );

    if( results == null ) return "";

jhaus

Options +FollowSymLinks
RewriteEngine on

RewriteCond %{THE_REQUEST} ^.*\/index\.html? 
RewriteRule ^(.*)index\.html?$ http://example.com$1 [R=301,L]

# More resources:
# 
# http://old.justinshattuck.com/2007/01/19/apache-mod_rewrite-cheat-sheet/
# http://net.tutsplus.com/tutorials/other/a-deeper-look-at-mod_rewrite-for-apache/