Ben bshyong

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.

Vim Manipulation Cheat Sheet

Action

Key	Result
`v`	select
`y`	copy (yank)
`c`	change
`d`	delete

Install ffmpeg

brew install ffmpeg

Download file through url, like this:

ffmpeg -protocol_whitelist file,http,https,tcp,tls,crypto -i "http://url-file.domain.m3u8" -c copy video.mp4

	#if UNITY_EDITOR
	using UnityEngine;
	#if UNITY_EDITOR
	using UnityEditor;
	using UnityEditor.Build;
	#endif

	public class BoardSetup : MonoBehaviour {
	[MenuItem("Board/Project Setup/Run All - Including Optional")]
	private static void RunAll() {

	{
	"id": "e75b8d14-6c1e-4984-840d-ea260fda6759",
	"revision": 0,
	"last_node_id": 1698,
	"last_link_id": 3003,
	"nodes": [
	{
	"id": 759,
	"type": "GetNode",
	"pos": [13560, 470],

	You are Manus, an AI agent created by the Manus team.

	You excel at the following tasks:
	1. Information gathering, fact-checking, and documentation
	2. Data processing, analysis, and visualization
	3. Writing multi-chapter articles and in-depth research reports
	4. Creating websites, applications, and tools
	5. Using programming to solve various problems beyond development
	6. Various tasks that can be accomplished using computers and the internet

	import argparse
	import json
	import logging
	import os
	import re
	import shutil
	from concurrent.futures import ProcessPoolExecutor, as_completed
	from dataclasses import dataclass
	from datetime import datetime
	from typing import Any, Callable, Dict, List, Literal, Optional, Tuple


	/*
	the twitter api is stupid. it is stupid and bad and expensive. hence, this.

	Literally just paste this in the JS console on the bookmarks tab and the script will automatically scroll to the bottom of your bookmarks and keep a track of them as it goes.

	When finished, it downloads a JSON file containing the raw text content of every bookmark.

	for now it stores just the text inside the tweet itself, but if you're reading this why don't you go ahead and try to also store other information (author, tweetLink, pictures, everything). come on. do it. please?
	*/

	#!/bin/bash
	# This script takes a PDF or list of PDFs and outputs a file(s)
	# named <file>_scanned.pdf that looks like it has been scanned
	#
	# Requires imagemagic and popper to be installed (brew install imagemagick poppler)
	#
	# Accepts: a list of files
	# Usage: ./<scriptfilename> pdf1.pdf pdf2.pdf
	#
	# To use as a macOS automator quick action you need to:

	Attribute VB_Name = "Newton"
	'Newton Utilities, written by swyx
	'Project initiated Jan 1 2012
	'V1 released Feb 5 2012
	'V1.1 added and renamed functions, updated helpfiles Apr 8 2012
	'V1.2 spline interp extraction from surface, n_RandomWalk. released Jun 15 2012
	'V1.21 n_Corr, n_TangencyPortfolio, n_CovMat subsumed into n_Cov, fixed n_GetTimeSeries, n_AutoCorr, n_Lag, n_Pval, added PValues to n_Regress, n_Payback, n_ChartMakeScatter
	'V1.22 n_Granger, pval for n_Corr, n_Divide, n_Curve, n_PriceIRS, upgraded n_BuildCurve
	'V1.23 n_PriceFwd/Rate, n_AutoRegress/n_Remove_AutoRegress (needs work), augmented n_PriceOption for pricing FXO, digitals, and returning greeks
	'V1.24 n_FXOVolCurve, n_FormatAsPercent linked to ctrl+shift+5, n_FormatInc/DecreaseDecimalPlace