hushicai hushicai

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.

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.

从零搭建一个高性能的前端CI服务器

本系列教程介绍如何从零搭建一个前端CI服务器，以及如何优化其性能。

本系列教程均基于Gitlab CI，其它系统的酌情参考。

[12:03 AM] acemarke: "controlled" and "uncontrolled" inputs
[12:04 AM] acemarke: if I have a plain, normal HTML page, and I put <input id="myTextbox" type="text" /> in my page(edited)
[12:04 AM] acemarke: and I start typing into that textbox
[12:04 AM] acemarke: it remembers what I've typed. The browser stores the current value for that input
[12:05 AM] acemarke: and then sometime later, I can get the actual element, say, const input = document.getElementById("myTextbox"), and I can ask it for its value: const currentText = input.value;
[12:05 AM] acemarke: good so far?
[12:08 AM] acemarke: I'll keep going, and let me know if you have questions
[12:08 AM] lozio: ok, actually I'm reading
[12:09 AM] lozio: good
[12:09 AM] acemarke: so, a normal HTML input field effectively stores its own value at all times, and you can get the element and ask for its value

What forces layout / reflow

All of the below properties or methods, when requested/called in JavaScript, will trigger the browser to synchronously calculate the style and layout*. This is also called reflow or layout thrashing, and is common performance bottleneck.

Generally, all APIs that synchronously provide layout metrics will trigger forced reflow / layout. Read on for additional cases and details.

Element APIs

Getting box metrics

elem.offsetLeft, elem.offsetTop, elem.offsetWidth, elem.offsetHeight, elem.offsetParent

The introduction to Reactive Programming you've been missing

(by @andrestaltz)

This tutorial as a series of videos

If you prefer to watch video tutorials with live-coding, then check out this series I recorded with the same contents as in this article: Egghead.io - Introduction to Reactive Programming.

Web Performance Power Tool: HTTP Archive (HAR) - blog post with video and details about the projects we covered.

The HAR Show GDL episode on YouTube

HAR Spec: http://www.softwareishard.com/blog/har-12-spec/
HTTP Archive (powered by HAR files): www.httparchive.org
HAR Viewer: http://code.google.com/p/harviewer/

	RN < 0.50 - watchman watch-del-all && rm -rf $TMPDIR/react-* && rm -rf node_modules/ && npm cache clean && npm install && npm start -- --reset-cache
	RN >= 0.50 - watchman watch-del-all && rm -rf $TMPDIR/react-native-packager-cache-* && rm -rf $TMPDIR/metro-bundler-cache-* && rm -rf node_modules/ && npm cache clean && npm install && npm start -- --reset-cache
	RN >= 0.63 - watchman watch-del-all && rm -rf node_modules && npm install && rm -rf /tmp/metro-* && npm run start --reset-cache

	npm >= 5 - watchman watch-del-all && rm -rf $TMPDIR/react-* && rm -rf node_modules/ && npm cache verify && npm install && npm start -- --reset-cache

	Windows - del %appdata%\Temp\react-native-* & cd android & gradlew clean & cd .. & del node_modules/ & npm cache clean --force & npm install & npm start -- --reset-cache

	<!DOCTYPE html>
	<html>
	<head>
	<meta charset=utf-8 />
	<title>JS Bin</title>
	</head>
	<body>
	<h2>Original CSS</h2>
	<style style="display: block; white-space: pre; font-family: monospace">
	h2 { margin:0; }

	Drop in replace functions for setTimeout() & setInterval() that
	make use of requestAnimationFrame() for performance where available
	http://www.joelambert.co.uk

	Copyright 2011, Joe Lambert.
	Free to use under the MIT license.
	http://www.opensource.org/licenses/mit-license.php