Skip to content

Instantly share code, notes, and snippets.

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.

---
description:
globs:
alwaysApply: true
---
# Code Guidelines
**Code Block Structure for Proposing Edits**:
When proposing code changes to be applied with the `edit_file` tool, the primary goal is to clearly specify the **complete target state** of the code block being modified. Represent unchanged surrounding lines using `// ... existing code ...` (or the equivalent comment syntax for the target language).
@ScriptedAlchemy
ScriptedAlchemy / CursorTools.json
Created January 31, 2025 03:54
Reverse Engineering cursor prompts
{
"tools": [
{
"type": "function",
"function": {
"name": "codebase_search",
"description": "Find snippets of code from the codebase most relevant to the search query.\nThis is a semantic search tool, so the query should ask for something semantically matching what is needed.\nIf it makes sense to only search in particular directories, please specify them in the target_directories field.\nUnless there is a clear reason to use your own search query, please just reuse the user's exact query with their wording.\nTheir exact wording/phrasing can often be helpful for the semantic search query. Keeping the same exact question format can also be helpful.",
"parameters": {
"type": "object",
"properties": {
@0xdevalias
0xdevalias / reverse-engineering-macos.md
Last active April 20, 2026 19:11
Some notes, tools, and techniques for reverse engineering macOS binaries
@pich4ya
pich4ya / proxychains-ng_m1.txt
Created March 6, 2023 03:25
Install proxychains-ng on macOS m1/m2 arm64e natively without Rosetta 2 (2023)
# @author Pichaya Morimoto ([email protected])
Problem:
```bash
brew install proxychains-ng
proxychains4 ncat 1.2.3.4 # not working
```
There are public workarounds like https://benobi.one/posts/running_brew_on_m1_for_x86/
@unixfox
unixfox / howto.md
Last active April 19, 2026 23:17
Install Alpine Linux on Oracle Cloud ARM VPS with Ubuntu pre-installed
@unixfox
unixfox / README.md
Last active May 3, 2025 08:57
Install alpine linux on Scaleway stardust
@alexalouit
alexalouit / README
Last active September 19, 2022 07:17
run alpine linux as ramdisk/iso/usb with zfs modules (modloop)
$ mkdir /tmp/a
$ cd /tmp/a
$ unsquashfs /media/sd**/boot/modloop-lts
$ mv squashfs-root/ lib
$ tar -xzvf /etc/apk/cache/zfs-lts-*.apk
$ depmod -b /tmp/a
$ mksquashfs lib/ modloop-lts -noappend -always-use-fragments
$ mount -o rw,remount /media/sd**
# do backup but not as /filename or /boot/filename, alpine will be use it)
$ mv /tmp/a/modloop-lts /media/sd**/boot/modloop-lts