Discover gists

CODEBASE REASONING TOPOLOGY (Short)

You are a thinking partner for experienced developers. Your role is to help them think clearer, design better systems, and ship coherent code — not to teach or act as a blind code generator.

Core Truth: Structure is persistence. Prioritize tight topology over perfect context.

Map the relationships deeply, even if you don't see the whole universe.

ENTRY PROTOCOL: Ambiguity Detection

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.

You are an expert software engineer and offensive security practitioner conducting a security audit of this codebase. Your goal is to find realistic, high-impact bugs and prove them against a running instance of the service.

Scope of interest (critical / high severity only)

Remote code execution
User impersonation / takeover
Authentication / authorization bypass
Detection / filter / trait bypass (if the service is a detection system)
Path traversal, SSRF, deserialization flaws, injection (SQL/command/log/header)
SQL injection, command injection, XSS injection and so on.

First we start by creating a wineprefix and installing our prerequisites from terminal:

WINEARCH=win32 WINEPREFIX=/home/$USER/GarminExpress winetricks dotnet452 vcrun2010 corefonts

WINEARCH=win32 WINEPREFIX=/home/$USER/GarminExpress winetricks win7

	const axios = require("axios")

	async function getCoordinates(city) {
	try {
	const { data } = await axios.get(
	"https://nominatim.openstreetmap.org/search",
	{
	params: {
	q: city,
	format: "json",

	Links from https://www.apple.com/final-cut-pro/trial/
	Final Cut Pro release notes https://support.apple.com/en-us/HT201237
	What's new in Final Cut Pro https://support.apple.com/en-us/HT207877
	Release history https://web.archive.org/web/20221121093911/https://en.wikipedia.org/wiki/Final_Cut_Pro_X#Release_history

	Final Cut Pro 10.1.3
	https://secure-appldnld.apple.com/Final_Cut_Pro_X/031-05296.20140814.VRR4r/FinalCutProTrial10.1.3.dmg

	Final Cut Pro 10.1.4
	https://secure-appldnld.apple.com/Final_Cut_Pro_X/031-02980.20141202.Jgt44/FinalCutProTrial10.1.4.dmg

	#!/usr/bin/swift

	// DISCLAIMER

	// This script modifies an unencrypted file associated with the trial version of Final Cut Pro.
	// Under the DMCA (17 U.S.C. § 1201), this modification does not qualify as circumvention of a technological
	// protection measure (TPM), as it does not involve bypassing encryption, authentication, or similar protections.
	// Distributing this code is therefore legal under the DMCA.

	// This script is intended for educational and research purposes, such as exploring trial-related file structures,

	const axios = require("axios")
	const crypto = require("crypto")

	const BASE = "https://www.perplexity.ai"

	const header_dasar = {
	"User-Agent":
	"Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/145.0.0.0 Safari/537.36",
	Accept: "text/event-stream",
	"Accept-Language": "id-ID,id;q=0.9,en-US;q=0.8,en;q=0.7",