karpathy

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.

gsarig

<?php 
/**
* Get Instagram media on WordPress using the current Instagram (Facebook) API
*
* @param $token // Info on how to retrieve the token: https://www.gsarigiannidis.gr/instagram-feed-api-after-june-2020/
* @param $user // User ID can be found using the Facebook debug tool: https://developers.facebook.com/tools/debug/accesstoken/
* @param int $limit // Add a limit to prevent excessive calls.
* @param string $fields // More options here: https://developers.facebook.com/docs/instagram-basic-display-api/reference/media
* @param array $restrict // Available options: IMAGE, VIDEO, CAROUSEL_ALBUM
*

gsarig

<?php
    $form_id = 3; 
    add_filter( 'gform_form_post_get_meta_' . $form_id, 'add_repeater_field' ); 
    function add_repeater_field( $form ) { 
      
      $field_id = 1000;
      $fields = [
        GF_Fields::create(
            [
              'type'       => 'text',

gsarig

#!/bin/bash

# A bash script to batch backup a server's WordPress databases with WP-CLI. It requires WP-CLI, obviously (https://wp-cli.org/).
# The script will search for all WordPress installations under a given directory and will backup their databases. You can declare more than one such directories.
# You don't have to modify the script every time you add a new site. As long as the new site is under a declared parent directory, its database will be backed up.
# After creating the file, don't forget to make it executable by doing:
# chmod +x wp-cli_batch_db_backup.sh
# Also, you might want to add this in a cron job for scheduled backups. 

# Set PATH environment variable

paceaux

Tiny rules for how to name stuff

CSS

How to name CSS classes

Stateful Class names

Is it a state that is only one of two conditions? (i.e. a boolean)

rileypaulsen

function wp_api_encode_acf($data,$post,$context){
	$data['meta'] = array_merge($data['meta'],get_fields($post['ID']));
	return $data;
}

if( function_exists('get_fields') ){
	add_filter('json_prepare_post', 'wp_api_encode_acf', 10, 3);
}

arunk-s

# backup.sh
# The contents of this file are released under the GNU General Public License. Feel free to reuse the contents of this work, as long as the resultant works give proper attribution and are made publicly available under the GNU General Public License.
# By Arun Sori <arunsori94@gmail.com>

#For taking backup of the desired directory and store it at a remote place

#timestamp
time_stamp=`date`

#backup file name 

mourner

// Lefalet shortcuts for common tile providers - is it worth adding such 1.5kb to Leaflet core?

L.TileLayer.Common = L.TileLayer.extend({
	initialize: function (options) {
		L.TileLayer.prototype.initialize.call(this, this.url, options);
	}
});

(function () {