Charley Wu akunzai

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.

Context Compaction

Research on how other coding assistants implement context compaction to manage long conversations.

Overview

Context compaction (also called "handoff" or "summarization") is a technique to manage the context window in long coding sessions. When conversations grow too long, performance degrades and costs increase. Compaction summarizes the conversation history into a condensed form, allowing work to continue without hitting context limits.

Claude Code

Setup an Always-on VSCode Tunnel on Oracle Cloud Always-free Instance

Access Oracle Cloud VM from any browser on any devices

Install VSCode for arm64

wget -O vscode.deb https://code.visualstudio.com/sha/download\?build\=stable\&os\=linux-deb-arm64 && sudo apt install ./vscode.deb --fix-broken -y && rm vscode.deb

Model 層級管理

Domain Model 層的意義

一個傳統 Java Spring 的應用程式會分為三層 @Controller @Service @Repository，然後依據組織需求，再往下細分。比方說 Service 層複雜的話，會增加 Gateway、Facade 層，或是我們組織獨有的 CoreService 層

另一個層級大部份 Java 組織會忽略的是 Domain Model 層，絕大多數 Java 開發者會開個 entity 或 model package，然後將所有的 Entity 往裡面丟。然而該 Entity 只是一堆欄位和 getter/setter 的堆疊，也就是單純的 Table 的對應物件而已， 不具備任何 business logic

Podman as a Docker Desktop alternative

Prerequisites

Install Homebrew from https://brew.sh

Install Podman

$ brew install podman

	#cloud-config
	package_upgrade: true

	ssh_authorized_keys:
	- <your key>

	packages:
	- apt-transport-https
	- ca-certificates
	- curl

	/**
	* Postman Pre-Request script to append CSRF token in header for POST requests in Laravel
	* Sanctum authenticated SPA. Requires active environment with {{url}} variable defined
	* for main app domain.
	*
	* Postman Interceptor allows appending cookies from browser, but Laravel CSRF middleware
	* only validates CSRF in headers or in _token form field, not in cookies. Axios automatically
	* appends the CSRF from cookie to headers, but Postman cannot access intercepted cookies
	* and use them, so we have to do one pre-request to get the CSRF token, store it
	* in environment so it can be reused, and then append it to headers.

	## HOMEBREW
	# Initialize Homebrew and set core PATHs
	if [[ $(uname -m) == 'arm64' ]] && [[ -x /opt/homebrew/bin/brew ]]; then
	eval "$(/opt/homebrew/bin/brew shellenv)"
	elif [[ -x /usr/local/bin/brew ]]; then
	eval "$(/usr/local/bin/brew shellenv)"
	fi

	## EXTERNAL TOOLS
	# OrbStack

	# This goes in your Microsoft.PowerShell_profile.ps1 (can find the path via $PROFILE in your prompt)
	Import-Module -Name posh-git,oh-my-posh,Terminal-Icons
	Set-PoshPrompt -Theme craver