Skip to content

Instantly share code, notes, and snippets.

Discover gists

@alwaysontheice
alwaysontheice / llm-wiki.md
Created September 7, 2026 03:16 — forked from karpathy/llm-wiki.md
llm-wiki

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.

@vimagick
vimagick / 评书MP3打包下载.tsv
Last active September 7, 2026 03:15
文件名 下载链接 提取码
一凡评书《孙庞演义》MP3免费打包下载 123回全 https://pan.baidu.com/s/1TKRhydXv0QLRiRcHSpBfWg?pwd=2233 2233
一凡评书《辽西风雨路》MP3打包下载 220回全 https://pan.baidu.com/s/1gqssQppHCpjWjeK9iqC31g?pwd=5566 5566
于枢海评书《大刀王五》MP3免费打包下载 79回全集 https://pan.baidu.com/s/12fIxuA4oI_zTVndaFpxqbA xdqt
于海龙评书《张义传奇》MP3免费打包下载 14回 https://pan.baidu.com/s/1viC2m6xWt7kVuww0rMdQ5g?pwd=6677 6677
仲维维评书《一个女人的风景》MP3免费打包下载 7回全集 https://pan.baidu.com/s/1c227Iuc wpja
仲维维评书《一滴血》MP3免费打包下载 24回全集 https://pan.baidu.com/s/1miG45L2 xm2j
仲维维评书《七星龙王》MP3免费打包下载 30回全集 https://pan.baidu.com/s/1mhEmtmw a3wx
仲维维评书《三少爷的剑》MP3打包下载 ed2k://|file|%5B%E4%BB%B2%E7%BB%B4%E7%BB%B4%5D%E4%B8%89%E5%B0%91%E7%88%B7%E7%9A%84%E5%89%91.rar|265302041|c210a0ac46be8f474738351d8d586bc8|h=MDB2DETOEWWKIT3NKIXDCZYKYKF4DJOH|/
仲维维评书《中国瓷商》MP3打包下载 75回全集 https://pan.baidu.com/s/1wX_Hx303j5jVP9xVMS6h5Q abah
@karpathy
karpathy / llm-wiki.md
Created April 4, 2026 16:25
llm-wiki

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.

@alexkar598
alexkar598 / ss13 on linux.md
Last active September 7, 2026 03:07
ss13 on linux

  1. Use ubuntu (or debian works too for this guide) because i dont know the other distros lmao

  2. Run sudo dpkg --add-architecture i386 to enable 32 bits package lists

  3. Run sudo apt update to make sure package lists are updated

  4. Run sudo apt install libstdc++6:i386 to get the dependencies for BYOND

  5. Find the download link for BYOND's linux version from http://www.byond.com/download/, looks like http://www.byond.com/download/build/513/513.1536_byond_linux.zip

@ivaniskandar
ivaniskandar / ScreenTransition.kt
Last active September 7, 2026 03:06
Predictive Back gesture support for Voyager (Jetpack Compose 1.7.0-alpha05+)
import androidx.activity.BackEventCompat
import androidx.activity.compose.PredictiveBackHandler
import androidx.compose.animation.AnimatedContent
import androidx.compose.animation.AnimatedContentTransitionScope
import androidx.compose.animation.ContentTransform
import androidx.compose.animation.EnterTransition
import androidx.compose.animation.ExitTransition
import androidx.compose.animation.SizeTransform
import androidx.compose.animation.core.AnimationSpec
import androidx.compose.animation.core.ExperimentalTransitionApi
@ave9858
ave9858 / Telemetry.md
Last active September 7, 2026 03:04

Easy Telemetry Disable for 10/11

Prerequisites

  1. Upgrade to Enterprise edition of Windows 10/11 if you are running Home or Pro.
    • You can do this through the Change Edition option in the Extras menu in MAS.

Disabling Telemetry via Group Policy

  1. Open the Group Policy Editor. Search for "Edit Group Policy" in search or run gpedit.msc.
@dctanner
dctanner / SKILL.md
Created July 6, 2026 09:47
Code in an Opus session, using Fable subagent to plan and Codex to write code and finally Fable subagent review for every task.
name cook
description Code in an Opus session, using Fable subagent to plan and Codex to write code and finally Fable subagent review for every task.
user-invokable true

Implement all tasks given to you. Use a Fable high subagent to write a plan.md for each task. Require my approval for all plans before beginning implementation. Then write all code by calling out to codex exec for each task using codex --yolo exec -c model="gpt-5.5" -c model_reasoning_effort="high" < plan.md (allow cmd to run for up to 20 minutes), then review code (using a subagent with Fable high) and smoke testing new features (usage a subagent with Opus medium) by running pnpm dev and doing real requests, before committing and pushing, then moving onto next task. Remember that every subagent or codex session is completely independent and must be given the full context it needs to complete its work.

@aamiaa
aamiaa / CompleteDiscordQuest.md
Last active September 7, 2026 03:00
Complete Recent Discord Quest

Caution

As of August 26th 2026, Discord has started suspending quest access of people caught automating quest completion.

Some users have received the following system message:

image

There isn't much I can do to make the script undetected, so use it at your own risk, as you WILL get flagged by doing so.

Complete Recent Discord Quest

@tm4dev
tm4dev / jk-bms-docs.md
Created August 27, 2025 18:22