Discover gists

LLM Wiki v2

A pattern for building personal knowledge bases using LLMs. Extended with lessons from building agentmemory 20K+ Stars ⭐️, a persistent memory engine for AI coding agents.

This builds on Andrej Karpathy's original LLM Wiki idea file. Everything in the original still applies. This document adds what we learned running the pattern in production: what breaks at scale, what's missing, and what separates a wiki that stays useful from one that rots.

What the original gets right

The core insight is correct: stop re-deriving, start compiling. RAG retrieves and forgets. A wiki accumulates and compounds. The three-layer architecture (raw sources, wiki, schema) works. The operations (ingest, query, lint) cover the basics. If you haven't read the original, start there.

tmux cheatsheet

As configured in my dotfiles.

start new:

tmux

start new with session name:

name

japanese-tech-writing

description

日本語の技術文書・書籍原稿の文章規範。整形（一文一行、引用ブロック、脚注、コラム記法）、段落と論証の構成（パラグラフライティング）、論証の厳密さ（ツッコミどころの除去）、読み手の負荷の管理、視点と語り、演出の抑制、LLM っぽい空句の禁止、冗長の排除を定める。日本語で技術書の章、草稿、記事、解説文を書くとき、または推敲・リライトするときに使用する。

日本語技術文書の文章規範

日本語で技術的な原稿（書籍の章、記事、解説文）を書く・推敲するときは、以下の規範に従う。

整形

Building a react native app in WSL2

Install, build and debug a react native app in WSL2 (Windows Subsystem for Linux) and Ubuntu.

Install tools in Windows
Enable mirrored networking mode in WSL2
Install tools in WSL2
Connect to android hardware device from Windows
Connect to android hardware device from WSL2
Connect to android virtual device in Windows

Download Wistia videos

right-click on the playing video, select Copy link
find Wistia video ID in the copied link e.g. wvideo=tra6gsm6rl
- alternative: look for e.g. hashedId=tra6gsm6rl in the page source
load http://fast.wistia.net/embed/iframe/ + video ID in your browser
look for "type":"original" in the page source and copy the URL from the next line e.g. "url":"http://embed.wistia.com/deliveries/129720d1762175bcd8e06dcab926ec76ad38ff00.bin"

alternative: look for "type":"hd_mp4_video"

Activate Windows 8, 8.1, 10 and 11 Pro for Free

A guide how to get and activate Windows 8, 8.1, 10 and 11 Pro for free!

NOTE

If you see the Windows keyboard button in this guide; and you can't find it on your keyboard, you likely have/had Windows 10 which has the button . If you can't find that one, you likely have a PC that has been upgraded to Windows 8/8.1/10/11 from Windows 8.1/8/7/Vista/XP and other ones. If you have one of those, refer the Windows key button to as yours. A list of them is below:

Windows key buttons

- Windows 11

- Windows 10

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/bash
	currentAttempt=0
	totalAttempts=10
	delay=15

	while [ $currentAttempt -lt $totalAttempts ]
	do
	currentAttempt=$(( $currentAttempt + 1 ))

	echo "Attempt $currentAttempt of $totalAttempts..."