Skip to content

Instantly share code, notes, and snippets.

View mtomas7's full-sized avatar

Tomas M. mtomas7

View GitHub Profile

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.

@dabit3
dabit3 / you_couldve_invented_openclaw.md
Last active May 9, 2026 00:03
You Could've Invented OpenClaw

See more of my writing here. Also check out Devin

In this post, I'll start from scratch and build up to OpenClaw's architecture step by step, showing how you could have invented it yourself from first principles, using nothing but a messaging API, an LLM, and the desire to make AI actually useful outside the chat window.

End goal: understand how persistent AI assistants work, so you can build your own (or become an OpenClaw power user).

First, let's establish the problem

When you use ChatGPT or Claude in a browser, there are several limitations:

@ScottKillen
ScottKillen / print.css
Last active March 7, 2026 14:30
PDF export stylesheet for obsidian
/* Obsidian snippet to style output of pdf exports
*/
@media print {
/* set your preferred fonts here.
*/
:root {
--body-font-family: "Newsreader Text", TimesNewRoman, "Times New Roman", Times, Baskerville, Georgia, serif;
@jakejackson1
jakejackson1 / disable-w3-total-cache-nags.php
Last active December 22, 2017 00:56
Remove Support Nag W3 Total Cache
<?php
/*
Plugin Name: Disable W3 Total Cache Nags
Version: 1.0
Author: Blue Liquid Designs
*/
add_action( 'wp_loaded', 'bld_disable_w3_total_cache_nags' );
function bld_disable_w3_total_cache_nags()
@gregrickaby
gregrickaby / html5-schema.org-markup.html
Last active August 2, 2022 00:05
Proper SCHEMA.ORG markup
<!DOCTYPE html>
<html lang="en-US">
<head>
<meta charset="UTF-8" />
<meta name="viewport" content="width=device-width, initial-scale=1" />
<meta name="robots" content="noodp, noydir" />
<link rel="dns-prefetch" href="//cdnjs.cloudflare.com">
<link rel="canonical" href="http://mysite.com/" />
<link rel="stylesheet" href="http://mysite.com/style.css" type="text/css" />
@thefuxia
thefuxia / all-actions-list.php
Created March 5, 2012 16:36
Plugin All Actions List
<?php # -*- coding: utf-8 -*-
/*
Plugin Name: All Actions List
Description: Lists all actions run during one request.
Version: 1.0
Required: 3.1
Author: Thomas Scholz
Author URI: http://toscho.de
License: GPL
*/
@Rarst
Rarst / r-debug.php
Last active June 12, 2025 01:56
R Debug (set of dump helpers for debug)
<?php
/*
Plugin Name: R Debug
Description: Set of dump helpers for debug.
Author: Andrey "Rarst" Savchenko
Author URI: https://www.rarst.net/
License: MIT
*/