Note
Obligatory disclaimer: this is for educational purposes only. I am not responsible for any damages caused by following this guide, or using any of the script(s) herein.
This guide prioritizes arm64 macOS, but may also work for other platforms.
| """ | |
| The most atomic way to train and run inference for a GPT in pure, dependency-free Python. | |
| This file is the complete algorithm. | |
| Everything else is just efficiency. | |
| @karpathy | |
| """ | |
| import os # os.path.exists | |
| import math # math.log, math.exp |
| local Library = loadstring(game:HttpGet("https://raw.githubusercontent.com/xHeptc/Kavo-UI-Library/main/source.lua"))() | |
| local Window = Library.CreateLib("Kiwi X V.2 ", "Serpent") | |
| local Tab = Window:NewTab("Aimlock") | |
| local Section = Tab:NewSection("Aimlock") | |
| Section:NewButton("Aimlock", "Aimlock", function() | |
| --[[ | |
| This source is free to use and hopefully a good source for new script developers to learn how to make an aimlock or silent aim. | |
| * If this is used in any script, please credit me for it. |
| #!/usr/bin/env python3 | |
| """ | |
| Scan a Linux panic QR code with ZXingReader and decompress its crash log. | |
| ./decode-kernel-panic.py screenshot.png -o panic.log | |
| """ | |
| import argparse | |
| import json | |
| from pathlib import Path | |
| import subprocess |
Note
Obligatory disclaimer: this is for educational purposes only. I am not responsible for any damages caused by following this guide, or using any of the script(s) herein.
This guide prioritizes arm64 macOS, but may also work for other platforms.
Written by Kaj Maney, April 2026 · Major revisions May and August 2026 A working Bluetooth page turner for the PW6 (MT8110 "Bellatrix") platform.
⚠️ Update — April 2026: An upstream fix is in progress that may significantly simplify this setup in future KOReader releases. @zampierilucas (the kindle-hid-passthrough author) has merged a uevent listener for input device hotplug into koreader-base. Once this lands in a KOReader release, UHID keyboards created by kindle-hid-passthrough may be detected natively, potentially allowing direct key-based page turning without the HTTP API workaround. If you're setting this up after the next KOReader release, test native key input first before following this guide. This guide remains valid for KOReader v2026.03 and earlier.
**
⚠️ Update — August 2026: the
| # author: eterna1_0blivion & soredake | |
| # Get a link to the Registration Entries file | |
| $link = "https://gist.github.com/eterna1-0blivion/70c1e5b14c7cfa8c6b6d574eb38fd27e/raw/context_pwsh_fix.reg" | |
| # Install them into Registry using PowerShell | |
| Invoke-WebRequest -Uri "$link" -OutFile "$env:TEMP/context_pwsh_fix.reg" | |
| reg import "$env:TEMP/context_pwsh_fix.reg" | |
| # Notify the user before exiting the program |
| { | |
| "accounts": [ | |
| { | |
| "profile": { | |
| "capes": [ | |
| ], | |
| "id": "49787d61f64b3b94a8f13d331a9ce421", | |
| "name": "XYZ", | |
| "skin": { | |
| "id": "", |
Purpose & scope. A defensive/educational knowledge base cataloguing known prompt-injection and jailbreak patterns, the models/systems they have affected, and the defenses against them. Compiled from primary literature (arXiv papers, vendor disclosures) and security research, June 2026.
How to read this. Every technique lists: how it works, an illustrative structural skeleton (the shape of the attack, not a weaponized payload), the models/systems it was reported against, and its current status. Examples are deliberately defanged. >