Skip to content

Instantly share code, notes, and snippets.

View ThatXliner's full-sized avatar
:atom:
Long live Atom

ThatXliner ThatXliner

:atom:
Long live Atom
View GitHub Profile

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.

#!/usr/bin/env bash
set -euo pipefail
# patch-claude-code.sh — Rebalance Claude Code prompts to fix corner-cutting behavior
#
# What this does:
# Patches the npm-installed @anthropic-ai/claude-code cli.js to rebalance
# system prompt instructions that cause the model to cut corners, simplify
# excessively, and defer complicated work.
#
"""
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
@Daeraxa
Daeraxa / pulsar-lang-comparison.md
Last active October 10, 2022 00:51
Comparison of default packages in Pulsar
@j1o1h1n
j1o1h1n / console.py
Last active February 11, 2025 18:34
This is a demo of an interactive console in a Textual user interface.
from __future__ import annotations
import string
import code
import sys
import io
from typing import Callable
from textual.app import App
from textual.widgets import Header, ScrollView
@melmsie
melmsie / cards.ts
Last active February 6, 2026 03:38
Dank Memer Blackjack Command Files
import * as Constants from './constants';
const randomInArray = <T>(arr: readonly T[]): T =>
arr[Math.floor(Math.random() * arr.length)];
export interface Card {
suit: typeof Constants.SUITS[number];
face: typeof Constants.FACES[number];
baseValue: number;
};
@sindresorhus
sindresorhus / esm-package.md
Last active May 5, 2026 02:50
Pure ESM package

Pure ESM package

The package that linked you here is now pure ESM. It cannot be require()'d from CommonJS.

This means you have the following choices:

  1. Use ESM yourself. (preferred)
    Use import foo from 'foo' instead of const foo = require('foo') to import the package. You also need to put "type": "module" in your package.json and more. Follow the below guide.
  2. If the package is used in an async context, you could use await import(…) from CommonJS instead of require(…).
  3. Stay on the existing version of the package until you can move to ESM.
@fnky
fnky / ANSI.md
Last active May 15, 2026 03:01
ANSI Escape Codes

ANSI Escape Sequences

Standard escape codes are prefixed with Escape:

  • Ctrl-Key: ^[
  • Octal: \033
  • Unicode: \u001b
  • Hexadecimal: \x1B
  • Decimal: 27
@DrSensor
DrSensor / Advanced Markdown Tricks.md
Last active March 6, 2026 09:30
Advanced Markdown Tricks

Repository

What Will I Learn?

In general, you will learn some markdown tricks combined with standard HTML tags. In more details what you will learn:

  • Hide-show content
  • Writing codeblocks inside codeblocks
  • Combining and using italic, bold, superscript, subscript, and/or strikethrough
  • Quoting long sentence (using nested blockquotes)
@nickoala
nickoala / 0_python_email.md
Last active March 2, 2026 05:58
Use Python to send and receive emails

Use Python to:

  • send a plain text email
  • send an email with attachment
  • receive and filter emails according to some criteria