ruohan.chen crhan

LLM Wiki v2

A pattern for building personal knowledge bases using LLMs. Extended with lessons from building agentmemory, 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.

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.

NUT Client for ESXi 6.0 Update 1 and below

Installation

NOTE: This package is not supported by VMWare and may cause problems in some cases like updating ESXi. So it is recommended to uninstall the VIB prior to major system changes. NOTE: NUT has its own official IANA port: 3493/TCP.

Download the binary: NutClient-ESXi500-1.3.0.tar.gz Source: NUT Official Downloads by René Garcia
Upload the binary to the Datastore

如何在路由器中实现透明代理？

0 互联网现状

目前整个互联网环境，被破坏最严重地部分，是 Web 服务体验。当直接破坏难以实现时，就会从流程链的上下游着手，如：DNS 污染。

其它地互联网服务类型，例如：邮件，可能小部分会受到 Web 服务上下游破坏地余震，但整体上基本不受影响。

	# Install tmux on rhel/centos 7

	# What do we want?
	libeventversion=2.1.11
	tmuxversion=3.1

	# install deps
	yum install gcc kernel-devel make ncurses-devel

	# DOWNLOAD SOURCES FOR LIBEVENT AND MAKE AND INSTALL

	# Run this as CoreOS user at home folder
	set -e
	toolbox dnf -y install bash-completion wget \
	&& toolbox wget https://raw.githubusercontent.com/docker/docker/master/contrib/completion/bash/docker -O /usr/share/bash-completion/completions/docker \
	&& toolbox cp /usr/share/bash-completion /media/root/var/ -R \
	&& source /var/bash-completion/bash_completion

	cp $(readlink .bashrc) .bashrc.new && mv .bashrc.new .bashrc

	echo "source /var/bash-completion/bash_completion" >> ~/.bashrc

	def execute(cmd, flag_file=None, msg_id='', debug=False):
	"""
	# http://stackoverflow.com/questions/1606795/catching-stdout-in-realtime-from-subprocess
	:param cmd: should never inlcude pipe or redirection, which would requires
	a new shell process
	This execute logs all stdout and stderr, which could look funny, especially
	when it comes to tools like aspc and wget
	"""
	logger.info('executing: {0}'.format(cmd))
	# todo: should check whether cmdsp includes pipe or redirection here

	# file: config/initialize/httparty.rb
	require 'httparty'

	module HTTParty
	class Request
	alias_method :_original_perform, :perform
	def perform(&block)
	payload = {
	method: http_method.const_get(:METHOD),
	url: uri

	#ruby '2.1.0'
	source 'https://rubygems.org'
	gem 'rack', '~>1.5.2'
	gem 'actionpack', '~> 4.1.0.beta1'
	gem 'railties', '~> 4.1.0.beta1'