Dan Borges ahhh

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.

ECB CryptoLocker Idea

Random musings from a 3 AM brain spin. Likely a very stupid idea :)

To Lock

create a list of files to lock
randomize the list order
generate a random 8 bit ascii hex string to use as an IV

	"""
	The most atomic way to train and run inference for a GPT in pure, dependency-free Python.
	This file is the complete algorithm.
	Everything else is just efficiency.

	@karpathy
	"""

	import os # os.path.exists
	import math # math.log, math.exp

	package main

	/*
	Example Go program with multiple .NET Binaries embedded

	This requires packr (https://github.com/gobuffalo/packr) and the utility. Install with:
	$ go get -u github.com/gobuffalo/packr/packr

	Place all your EXEs are in a "binaries" folder

	'''
	parse SavedState artifacts extracted from OSX.

	author: Willi Ballenthin (william.ballenthin@fireeye.com)
	license: Apache 2.0
	'''
	import re
	import sys
	import json
	import struct

	#!/usr/bin/env python2.7
	import argparse
	import binascii
	import sys
	import base64
	import hashlib
	from Crypto.Cipher import AES
	from pkcs7 import PKCS7Encoder
	import random
	from random import randint

	#!/usr/bin/env python3
	'''
	bling.py - extract keys from macOS keychains.

	installation:
	pip install pytz hexdump vivisect-vstruct-wb tabulate argparse pycryptodome

	usage:
	python bling.py /path/to/keychain-db <password> ./path/to/output/directory

	// Tracking cursor position in real-time without JavaScript
	// Demo: https://twitter.com/davywtf/status/1124146339259002881

	package main

	import (
	"fmt"
	"net/http"
	"strings"
	)

	'''
	parse osx sticky databases.

	author: Willi Ballenthin <william.ballenthin@fireeye.com>
	license: Apache 2.0

	usage:

	$ python extract_stickies.py /path/to/input.bin /path/to/output/directory/
	'''



	package main

	import (
	"io/ioutil"
	"os"
	"os/exec"
	"strconv"
	"syscall"