joest joest67

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.

	#!/bin/sh

	# Source config file
	# should contains follows variables:
	# AUTH_TOKEN
	# ROOM_ID
	source config.sh

	my_func() {
	return 1

	#! /usr/bin/env python

	""" Convert values between RGB hex codes and xterm-256 color codes.

	Nice long listing of all 256 colors and their codes. Useful for
	developing console color themes, or even script output schemes.

	Resources:
	* http://en.wikipedia.org/wiki/8-bit_color
	* http://en.wikipedia.org/wiki/ANSI_escape_code

	#!/bin/sh

	# source: https://github.com/florianl/AutoUpdateGit/blob/master/AutoUpdateGit.sh

	# Change here your basedirectory
	# Example:
	# Base="/this/is/my/basedirectory"
	Base=""

	if [ -z "$Base" -o "$Base" == "/this/is/my/basedirectory" ];

	430972,5037,6,52,2600.0,0,用,c
	352986,4498,5,12,600.0,0,对,-
	1399015,26982,569,12,600.0,0,-
	678560,2954497,11,9,450.0,1,用,扫
	331,18779,2,9,450.0,0,用,地址,-,win
	135200,3568182,2,6,300.0,1,端口,路由,转发
	992422,21469,36,6,300.0,0,脚本,存,window,调用
	142786,26725,3,6,300.0,0,网,地址,-
	281,18243,2,5,250.0,0,屏蔽,端口,路由,路由器
	701894,22114,12,4,200.0,0,用

	#!/bin/sh

	###
	# SOME COMMANDS WILL NOT WORK ON macOS (Sierra or newer)
	# For Sierra or newer, see https://github.com/mathiasbynens/dotfiles/blob/master/.macos
	###

	# Alot of these configs have been taken from the various places
	# on the web, most from here
	# https://github.com/mathiasbynens/dotfiles/blob/5b3c8418ed42d93af2e647dc9d122f25cc034871/.osx

	#! /usr/bin/env python3
	'''pyCookieCheat.py
	2015022 Now its own GitHub repo, and in PyPi.
	- For most recent version: https://github.com/n8henrie/pycookiecheat
	- This gist unlikely to be maintained further for that reason.
	20150221 v2.0.1: Now should find cookies for base domain and all subs.
	20140518 v2.0: Now works with Chrome's new encrypted cookies.

	See relevant post at http://n8h.me/HufI1w



	test:
	@export env=test; \
	echo $$env