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.

tomkrikorian

name	visionos-agent
description	Senior visionOS Engineer and Spatial Computing Expert for Apple Vision Pro development.

VISIONOS AGENT GUIDE

ROLE & PERSONA

You are a Senior visionOS Engineer and Spatial Computing Expert. You specialize in SwiftUI, RealityKit, and ARKit for Apple Vision Pro. Your code is optimized for the platform, adhering strictly to Apple's Human Interface Guidelines for spatial design.

igor-makarov

Replace Xcode MCP with CLI (saves about 32k tokens)

There are two files:

• xcodebuild - the script that the agent is expected to run
• building-with-xcode.md - the instructions given to the agent, i.e. in CLAUDE.md

kieranklaassen

You are an experienced software developer tasked with addressing a GitHub issue. Your goal is to analyze the issue, understand the codebase, and create a comprehensive plan to tackle the task. Follow these steps carefully:

1. First, review the GitHub issue using the gh issue view command.

<github_issue> #$ARGUMENTS </github_issue>

2. Next, examine the relevant parts of the codebase.

Analyze the code thoroughly until you feel you have a solid understanding of the context and requirements.

astashov

const { google } = require("googleapis");
const fs = require("fs");

// You need to go to Google Cloud console (console.cloud.google.com), then APIs
// & Services -> Credentials. There, create a new service account (or reuse
// existing if you have one). Click on a service account, go to Keys, and create
// a new key, and download JSON for it.  You'll use path to that JSON for
// SERVICE_ACCOUNT_FILE var.
//
// Then, go to Google Play Console, then "Users and Permissions" on the left

astashov

const fetch = require("node-fetch");
const fs = require("fs");
const jwt = require("jsonwebtoken");
const util = require("util");

// To get the private key, go to App Store Connect (appstoreconnect.apple.com), then "Users and Access"
// at the top. Then go to "Integrations" -> "App Store Connect API", under "Team Keys" create a new key,
// and you'll be able to download the private key for it.
const PRIVATE_KEY = fs.readFileSync("AuthKey_F2BLAHBLAH.p8", "utf8");

luthviar

class ViewController: UIViewController, BottomSheetPeopleDetectedInputDelegate {
    private var overlayView: UIView?
    
    override func viewDidLoad() {
        super.viewDidLoad()
        view.backgroundColor = .white
        
        let showButton = UIButton(type: .system)
        showButton.setTitle("Show Bottom Sheet", for: .normal)
        showButton.addTarget(self, action: #selector(showBottomSheet), for: .touchUpInside)

0xdevalias

Audio Pitch Correction (eg. autotune, melodyne, etc)

Some notes on Audio pitch correction (eg. autotune, melodyne, etc).

Table of Contents

• Melodyne
• FLStudio NewTone
• Synchronarts RePitch

martweiss

artistImageUrlList = []
print('Total artists streamed:', len(uniqueArtistUriList))
for uri in range(49):
    spotifyUrl = requests.get('https://open.spotify.com/artist/' +
        str(uniqueArtistUriList[uri]))
    spoifyHtmlData = str(soup(spotifyUrl.text, 'html.parser'))
    spoifyHtmlData = spoifyHtmlData.split('<')
    artistName = spoifyHtmlData[5].split('>')
    artistName = artistName[1].split('|')
    artistApiResults = requests.get('https://api.spotify.com/v1/search',

markusand

import winston from 'winston';
import morgan from 'morgan';
import json from 'morgan-json';

/* WINSTON */
const filterLevel = level => winston.format(info => info.level === level ? info : undefined)();

const logger = winston.createLogger({
  levels: { error: 0, warn: 1, info: 2, debug: 3, http: 4 },
  transports: [