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.

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.

This is a guide on how to activate ur idevice on ios 15 (maybe ios 14 idk) when its futurerestored on 16 sep.

DISCLAIMER: I am very aware people might use this to bypass icloud, but I am NOT encouraging you to bypass icloud.

ALSO: THIS IS RECOMMENDED FOR ADVANCED USERS ONLY, i am not responsable for ur device being broken (tho idk how it becomes broken via this)

also, i've only tested this on macOS, might work on linux tho if u replace darwin with linux and shit.

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.

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.

Методичка: от первоначальной идеи и полного ТЗ до проверенного локального приложения и контролируемого релиза.

Парный документ к методичке «Автономная продуктовая компания на базе Codex». Методология одна; здесь — её посадка на архитектуру и инструменты Claude Code. И в отличие от «сделайте сами» — эта посадка уже реализована и опубликована: Claude Code Starter v6.3.0 «Orchestrator Model».

Please consider using http://lygia.xyz instead of copy/pasting this functions. It expand suport for voronoi, voronoise, fbm, noise, worley, noise, derivatives and much more, through simple file dependencies. Take a look to https://github.com/patriciogonzalezvivo/lygia/tree/main/generative

float rand(float n){return fract(sin(n) * 43758.5453123);}

float noise(float p){
	float fl = floor(p);
  float fc = fract(p);