jabez007

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.

jabez007

Scan for Devices

bluetoothctl

sudo bluetoothctl
  scan on
  # wait for the device to show up
  scan off
  quit

BDADDR=<target device address>

jabez007

hex2ascii() {
    tr -d "'" | xxd -r -p; printf "\n"
}

gattval() {
    cut -d":" -f2 | tr -d "'" | xxd -r -p; printf "\n"
}

gatt_read() {
    if [ $# -lt 1 ]; then

jabez007

#!/bin/bash

CLIENT_ID='Iv23ctfURkiMfJ4xr5mv'
COMMON_HEADERS=(
  -H 'accept: application/json'
  -H 'content-type: application/json'
  -H 'accept-encoding: gzip,deflate,br'
)

# Step 1: Request device and user codes

jabez007

#!/bin/bash

# Check if container ID is provided
if [ -z "$1" ]; then
  echo "Usage: $0 <container_id_or_name> [sleep_interval_in_seconds]"
  exit 1
fi

# Container ID or name from the first argument
CONTAINER_ID=$1

jabez007

input price = FundamentalType.CLOSE;
input aggregationPeriod = AggregationPeriod.DAY;
input length = 9;
input displace = 0;
input showOnlyLastPeriod = no;

plot DailyEMA;

if showOnlyLastPeriod and !IsNaN(close(period = aggregationPeriod)[-1]) {
    DailyEMA = Double.NaN;

jabez007

You will now act as a title and description generator for products listed on an e-commerce platform called "Etsy". Esty hosts product listings for sale by independent crafters. I will provide you some basic examples of good product titles and descriptions found on Etsy.

Examples:

1. • Title: Set of 6 Bird Pun Coasters
   • Description:

   A collection of 6 coasters, each featuring a different bird chart from the best selling ‘For the Pun of It’ collection.
   

jabez007

You will now act as a prompt generator for a generative AI called "Leonardo AI". Leonardo AI generates images based on given prompts. I will provide you basic information required to make a Stable Diffusion prompt, you will never alter the structure in any way and obey the following guidelines.

Basic information required to make Leonardo AI prompt:

• Prompt structure:

  • Coloring Page Images prompt structure will be in this format "Subject Description in details with as much as information can be provided to describe image, Art Styles, Art Inspirations, Render Related Information"

  • Artistic Image Images prompt structure will be in this format "Type of Image, Subject Description, Art Styles, Art Inspirations, Camera, Shot, Render Related Information"

jabez007

using System;
using System.Linq;
using System.Threading.Tasks;

namespace Utils
{
  /// <summary>
  /// Expanding on the idea from https://alastaircrabtree.com/implementing-the-retry-pattern-for-async-tasks-in-c/
  /// </summary>
  public static class RetryHelper

jabez007

exports.RateLimit = (function () {
  var RateLimit = function (maxOps, interval, allowBursts) {
    this._maxRate = allowBursts ? maxOps : maxOps / interval;
    this._interval = interval;
    this._allowBursts = allowBursts;

    this._numOps = 0;
    this._start = new Date().getTime();
    this._queue = [];
  };