Carlos Granier cgranier

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.

Título del Proyecto

Acá va un párrafo que describa lo que es el proyecto

Comenzando 🚀

Estas instrucciones te permitirán obtener una copia del proyecto en funcionamiento en tu máquina local para propósitos de desarrollo y pruebas.

Mira Deployment para conocer como desplegar el proyecto.

	#!/bin/bash
	# Sets EXIF date on an image based on YYYY-MM-DD pattern in filename or manual date.
	set -euo pipefail

	show_help() {
	cat <<EOF
	Usage: $(basename "$0") [OPTIONS] <image>

	Sets EXIF date on an image file. Extracts date from YYYY-MM-DD pattern in
	filename, or uses manually specified date. Skips if EXIF date already exists.

	Project Approach

	* Always check for a PRD (Product Requirements Document) before starting a new task and follow it closely
	* Look for comprehensive project documentation to understand requirements before making changes
	* Focus only on code areas relevant to the assigned task
	* Prefer iterating on existing code rather than creating new solutions
	* Keep solutions simple and avoid introducing unnecessary complexity

	Code Quality

	// 1. Log in to your Amazon account
	// 2. Go to your Content Library > Books - https://www.amazon.com/hz/mycd/digital-console/contentlist/booksAll/dateDsc/
	// 3. Open your browser's Javascript console
	// 4. For each page of books, paste this script into the console

	(async function () {
	// Close the notification if it appears
	function closeNotification() {
	const notifClose = document.querySelector("span#notification-close");
	if (notifClose) {

	<!DOCTYPE html>
	<html>
	<head>
	<base target="_top">
	<title>Message Display Test</title>
	<link rel="stylesheet" href="https://ssl.gstatic.com/docs/script/css/add-ons1.css">
	</head>
	<body style="padding:3em;">
	<h1>Messages</h1>
	<ul>

	// Just before switching jobs:
	// Add one of these.
	// Preferably into the same commit where you do a large merge.
	//
	// This started as a tweet with a joke of "C++ pro-tip: #define private public",
	// and then it quickly escalated into more and more evil suggestions.
	// I've tried to capture interesting suggestions here.
	//
	// Contributors: @r2d2rigo, @joeldevahl, @msinilo, @_Humus_,
	// @YuriyODonnell, @rygorous, @cmuratori, @mike_acton, @grumpygiant,