Skip to content

Instantly share code, notes, and snippets.

View techtherapy's full-sized avatar

Gareth techtherapy

3 followers · 5 following
View GitHub Profile
@brandonb927
brandonb927 / osx-for-hackers.sh
Last active September 8, 2026 14:09
OSX for Hackers: Yosemite/El Capitan Edition. This script tries not to be *too* opinionated and any major changes to your system require a prompt. You've been warned.
#!/bin/sh
###
# SOME COMMANDS WILL NOT WORK ON macOS (Sierra or newer)
# For Sierra or newer, see https://github.com/mathiasbynens/dotfiles/blob/master/.macos
###
# Alot of these configs have been taken from the various places
# on the web, most from here
# https://github.com/mathiasbynens/dotfiles/blob/5b3c8418ed42d93af2e647dc9d122f25cc034871/.osx
@bryanzak
bryanzak / ARDFixer.command
Last active March 3, 2017 05:36
ARD Admin 3.7.1 Fixer - a script to cause ARD Admin to effectively "forget" all cached IP and DNS information for known computers. This deals with a problem in ARD Admin 3.7.1 where it seems to be too aggressive in caching this information
#!/bin/bash
SCRIPT_VERSION="1.0"
# some background info here: https://discussions.apple.com/message/24596710#24596710
# verify ARD Admin version
# quit ARD Admin - if running, wait 10 seconds
# kill cfprefsd
# rename plist
# sed
@arubdesu
arubdesu / makefile
Last active September 4, 2021 18:02
Luggage makefile-generated pkg to drop dummy receipts in place which stop prompt to update/adopt Mac AppStore apps (in this case for iLife without Garageband)(may also need to wipe /var/db/.MASManifest, as per @jessepeterson)
USE_PKGBUILD=1
include /usr/local/share/luggage/luggage.make
TITLE=iWorkDummyReceipts
REVERSE_DOMAIN=org.my.iWorkDummyReceipts
PACKAGE_VERSION=0.1
PAYLOAD=pack-iWorkDummyReceipts
pack-iWorkDummyReceipts: l_Applications
@sudo mkdir -p -m 755 ${WORK_D}/Applications/{Numbers,Pages,Keynote}.app/Contents/_MASReceipt
@sudo touch ${WORK_D}/Applications/{Numbers,Pages,Keynote}.app/Contents/_MASReceipt/receipt
@richardW8k
richardW8k / RW_Button_Field.php
Last active January 15, 2020 08:44
Adds a Button field type allowing you to add extra buttons to the form, supported type are button, submit and reset.
<?php
/**
* Plugin Name: Gravity Forms - Button Field
* Last Modified: 22/06/2014
*/
class RW_Button_Field {
function __construct() {
if( ! property_exists( 'GFCommon', 'version' ) || ! version_compare( GFCommon::$version, '1.8', '>=' ) )
@opshope
opshope / reveal_autologin_password.sh
Created November 17, 2015 18:02
Decrypt the auto-login password stored in /etc/kcpassword on OSX
sudo ruby -e 'key = [125, 137, 82, 35, 210, 188, 221, 234, 163, 185, 31]; IO.read("/etc/kcpassword").bytes.each_with_index { |b, i| break if key.include?(b); print [b ^ key[i % key.size]].pack("U*") }'
@talkingmoose
talkingmoose / Download and Install Microsoft product.zsh
Last active August 7, 2026 02:35
**Download the latest version of this script from: https://gist.github.com/b6637160b65b751824943ede022daa17 .** Downloads and installs the latest available Microsoft product specified directly on the client. This avoids having to manually download and store an up-to-date installer on a distribution server every month.
#!/bin/zsh
:<<'ABOUT_THIS_SCRIPT'
-----------------------------------------------------------------------
Written by:William Smith
Partner Program Manager
Jamf
bill@talkingmoose.net
https://gist.github.com/b6637160b65b751824943ede022daa17
@karpathy
karpathy / llm-wiki.md
Created April 4, 2026 16:25
llm-wiki

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.