0xdevalias

Reverse Engineering Webpack Apps

Some notes and techniques for reverse engineering Webpack (and a little bit about React/Vue/Angular) apps.

Table of Contents

• Webpack (and similar bundlers) Chunk Format
  • AST Structure
• Basic minified example

nathaningram

<?php
   /*
   Plugin Name: Custom Dashboard Widgets
   Plugin URI: https://wpnathan.com
   Description: Creates Custom Dashboard Widgets for Client Websites
   Version: 1.0
   Author: Nathan Ingram
   Author URI: https://wpnathan.com
   License: GPL2
   */

nathaningram

<?php
/*
Plugin Name: Custom Security Functions
Description: Customized WP Security
Version: 2024.11
Plugin URI: https://wpnathan.com
Author: Nathan Ingram
Author URI: https://wpnathan.com
License: GPL2
*/

nathaningram

<?php
   /*
   Plugin Name: Custom Media Handling Functions
   Plugin URI: https://wpnathan.com
   Description: Customize WordPress Default Media Behavior
   Version: 2024.11
   Author: Nathan Ingram
   Author URI: https://wpnathan.com
   License: GPL2
   */

boringmarketer

The Direct Response Copy Skill

Write copy that converts. Landing pages, emails, sales copy, headlines, CTAs, social posts — anything persuasive.

This is an AI skill file. It turns any AI into a direct response copywriter trained on the frameworks of Schwartz, Hopkins, Ogilvy, Halbert, Caples, Sugarman, and Collier. Instead of getting generic AI copy, you get internet-native writing that sounds like a smart friend explaining something — while quietly deploying every persuasion principle in the book.

---

How to use this

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.

kasajian

        HWND prevWindow = NULL;
        HWND hwnd;
        for ( ;; )
        {
            hwnd = FindWindowEx( HWND_MESSAGE, prevWindow, L"OleMainThreadWndClass", NULL );
            if ( !hwnd )
                break;

 if ( GetWindowThreadProcessId( hwnd, NULL ) == GetCurrentThreadId() )