M2kar m2kar

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.

How to add a collapsible section in markdown

1. Example

Click me

Heading

Foo
Bar
- Baz

Qux

Transparent proxy for Docker containers

If the Docker host is placed inside a proxy server, it needs to add the proxy configuration to each Dockerfile such as ENV http_proxy.

Following allows transparent access from the container to outside without any proxy configuration.

Set up the transparent proxy
Apply iptables rule for the transparent proxy

tmux shortcuts & cheatsheet

start new:

tmux

start new with session name:

tmux new -s myname

	// 自定义 compact 布局 — 所有面板无边框，最大化复制友好度
	layout {
	pane borderless=true
	pane size=1 borderless=true {
	plugin location="compact-bar"
	}
	}

	Windows Registry Editor Version 5.00

	[HKEY_CLASSES_ROOT\Directory\Background\shell\git_shell]
	@="Git Ba&sh Here"
	"Icon"="D:\\Git\\git-bash.exe"

	[HKEY_CLASSES_ROOT\Directory\Background\shell\git_shell\command]
	@="\"C:\\Users\\iMakar\\AppData\\Local\\Microsoft\\WindowsApps\\wt.exe\" -p GitBash -d \"%V\""

	[HKEY_CLASSES_ROOT\Directory\Background\shell\Debian]

	http://download.huzheng.org/dict.org/stardict-dictd_www.dict.org_gcide-2.4.2.tar.bz2
	http://download.huzheng.org/dict.org/stardict-dictd_www.dict.org_wn-2.4.2.tar.bz2
	http://download.huzheng.org/dict.org/stardict-dictd-moby-thesaurus-2.4.2.tar.bz2
	http://download.huzheng.org/dict.org/stardict-dictd-web1913-2.4.2.tar.bz2
	http://download.huzheng.org/dict.org/stardict-dictd_www.dict.org_elements-2.4.2.tar.bz2
	http://download.huzheng.org/dict.org/stardict-dictd-vera-2.4.2.tar.bz2
	http://download.huzheng.org/dict.org/stardict-dictd-jargon-2.4.2.tar.bz2
	http://download.huzheng.org/dict.org/stardict-dictd_www.dict.org_foldoc-2.4.2.tar.bz2
	http://download.huzheng.org/dict.org/stardict-dictd-easton-2.4.2.tar.bz2
	http://download.huzheng.org/dict.org/stardict-dictd_www.dict.org_hitchcock-2.4.2.tar.bz2

	# Ruby is our language as asciidoctor is a ruby gem.
	lang: ruby
	before_install:
	- sudo apt-get install pandoc
	- gem install asciidoctor
	script:
	- make
	after_success:
	- .travis/push.sh
	env:

	var request = require('request');
	var unzip = require('unzip');
	var csv2 = require('csv2');

	request.get('http://s3.amazonaws.com/alexa-static/top-1m.csv.zip')
	.pipe(unzip.Parse())
	.on('entry', function (entry) {
	entry.pipe(csv2()).on('data', console.log);
	})
	;

	#!/usr/bin/env python

	## Tiny Syslog Server in Python.
	##
	## This is a tiny syslog server that is able to receive UDP based syslog
	## entries on a specified port and save them to a file.
	## That's it... it does nothing else...
	## There are a few configuration parameters.

	LOG_FILE = 'youlogfile.log'