Damián Caruso dalecaru

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.

MacOS

Build 3059

MD5: 59bab8f71f8c096cd3f72cd73851515d

Rename it to: Sublime Text

Make it executable with: chmod u+x Sublime\ Text

	env:
	- tf_version=0.8.8
	- tf_version=0.9.11

	sudo: required

	language: bash

	before_install:
	- wget https://releases.hashicorp.com/terraform/${tf_version}/terraform_${tf_version}_linux_amd64.zip -O /tmp/terraform.zip

	import paho.mqtt.client as paho
	import os
	import socket
	import ssl
	from time import sleep
	from random import uniform
	import json

	import logging
	logging.basicConfig(level=logging.INFO)

	apt-get install -y libreadline-dev libncurses5-dev libpcre3-dev \
	libssl-dev perl make build-essential



	wget https://github.com/openresty/stream-lua-nginx-module/archive/v0.0.1.tar.gz \
	-O /tmp/stream-lua-nginx-module-0.0.1.tar.gz

	mkdir /tmp/stream-lua-nginx-module-0.0.1
	tar xzvf /tmp/stream-lua-nginx-module-0.0.1.tar.gz -C /tmp/stream-lua-nginx-module-0.0.1

	#!/bin/bash
	#
	# What is that
	# ============
	#
	# This script will help you setting up your digital ocean
	# infrastructure with Ansible v2.0+ and DO API v2
	#
	# Usually, when working with DO, one is supposed to use digital_ocean.py
	# inventory file, and spin up instances in a playbook.

	var gulp = require('gulp');
	var gutil = require('gulp-util');
	var del = require('del');
	var rename = require('gulp-rename');
	var install = require('gulp-install');
	var zip = require('gulp-zip');
	var AWS = require('aws-sdk');
	var fs = require('fs');
	var runSequence = require('run-sequence');

	{"address":"0x7f8b8df20628", "type":"STRING", "class":"0x7f8b8a029668", "frozen":true, "embedded":true, "fstring":true, "bytesize":3, "value":"=5F", "encoding":"UTF-8", "flags":{"wb_protected":true, "old":true, "marked":true}}
	{"address":"0x7f8b8df20650", "type":"STRING", "class":"0x7f8b8a029668", "frozen":true, "embedded":true, "fstring":true, "bytesize":3, "value":"=3F", "encoding":"UTF-8", "flags":{"wb_protected":true, "old":true, "marked":true}}
	{"address":"0x7f8b8df20678", "type":"STRING", "class":"0x7f8b8a029668", "frozen":true, "embedded":true, "fstring":true, "bytesize":3, "value":"=29", "encoding":"UTF-8", "flags":{"wb_protected":true, "old":true, "marked":true}}
	{"address":"0x7f8b8df206a0", "type":"STRING", "class":"0x7f8b8a029668", "frozen":true, "embedded":true, "fstring":true, "bytesize":3, "value":"=28", "encoding":"UTF-8", "flags":{"wb_protected":true, "old":true, "marked":true}}
	{"address":"0x7f8b8df206c8", "type":"STRING", "class":"0x7f8b8a029668", "frozen":true, "embedded":true, "fstring":tr

	#!/usr/bin/env ruby

	require 'ostruct'

	class LinuxProcess < OpenStruct

	def initialize(args)
	super
	read_status!
	end

	# Install Git needed for Git based gems
	packages:
	yum:
	git: []