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.

@burkeholland
burkeholland / ainstall.md
Last active November 8, 2026 16:16
Ultralight Orchestration

Ultralight Orchestration

A minimal multi-agent system with an orchestrator, a planner, a coder, and a designer working together providing orchestration between Claude, Codex and Gemini.

Instructions

Install all agents listed below into VS Code Insiders...

Title Type Description
@yifanzz
yifanzz / rules.md
Created March 27, 2025 20:59
YouTube Episode Supabase

Database

  • Use supabase for database queries and schema
  • Before performing any database related tasks, make sure to read the database.types.ts first for existing database schema
  • Always use migrations to update the database schema, create them using the command npx supabase migration new <migration-name>
  • Afer creating a migration file, run npx supabase migration up to apply the migration and run npx supabase gen types typescript --local > src/types/database.types.ts to generate the type file
  • When creating a new table, it must have columns for created_at and updated_at and the values should be set automatically via triggers using public.handle_created_at() and public.handle_updated_at()
  • Always enable Row Level Security (RLS) on newly create tables via ALTER TABLE <table_name> ENABLE ROW LEVEL SECURITY; in migration files and add reasonable policies
  • Always use await createServerClient() in the @/utils/supabase/server to create supabase client in server components and `createBrow
@dnoliver
dnoliver / Dockerfile
Last active October 12, 2020 12:47
Custom Application Using TPM PKCS11
FROM fedora:latest
RUN dnf install -y tpm2-pkcs11 tpm2-pkcs11-tools tpm2-tools gnutls-utils openssl-pkcs11 nodejs
VOLUME /etc/tpm2_pkcs11
VOLUME /root/test
WORKDIR /root/test
CMD bash
@chiro-hiro
chiro-hiro / golang-uint64-uint32-to-bytes.md
Last active October 1, 2024 09:04
Golang uint64, uint32 to bytes array

Wrote for a joyful journey of Go.

package main

import (
	"fmt"
)

func i64tob(val uint64) []byte {
@noamtamim
noamtamim / README.md
Last active June 2, 2026 15:19
Markdown with PlantUML

How to use PlantUML with Markdown

PlantUML is a really awesome way to create diagrams by writing code instead of drawing and dragging visual elements. Markdown is a really nice documentation tool.

Here's how I combine the two, to create docs with embedded diagrams.

Step 0: Setup

Get the command-line PlantUML from the download page or your relevant package manager.

@stewartadam
stewartadam / main.py
Last active August 30, 2026 12:50 — forked from gear11/main.py
Simple Python proxy server based on Flask and Requests with support for GET and POST requests.
"""
A simple proxy server, based on original by gear11:
https://gist.github.com/gear11/8006132
Modified from original to support both GET and POST, status code passthrough, header and form data passthrough.
Usage: http://hostname:port/p/(URL to be proxied, minus protocol)
For example: http://localhost:5000/p/www.google.com
"""
import re
@RichardBronosky
RichardBronosky / README.md
Last active February 24, 2025 04:59
Using cloud-init for cloudless provisioning of Raspberry Pi

Installing cloud-init on a fresh Raspbian Lite image

This is a work in Progress!

Purpose

This mainly demonstrates my goal of preparing a Raspberry Pi to be provisioned prior to its first boot. To do this I have chosen to use the same cloud-init that is the standard for provisioning servers at Amazon EC2, Microsoft Azure, OpenStack, etc.

I found this to be quite challenging because there is little information available for using cloud-init without a cloud. So, this project also servers as a demonstration for anyone on any version of Linux who may want to install from source, and/or use without a cloud. If you fall into that later group, you probably just want to read the code. It's bash so everything I do, you could also do at the command line. (Even the for loop.)

@posener
posener / go-table-driven-tests-parallel.md
Last active August 21, 2026 08:12
Be Careful with Table Driven Tests and t.Parallel()

Be Careful with Table Driven Tests and t.Parallel()

We Gophers, love table-driven-tests, it makes our unittesting structured, and makes it easy to add different test cases with ease.

Let’s create our table driven test, for convenience, I chose to use t.Log as the test function. Notice that we don't have any assertion in this test, it is not needed to for the demonstration.

func TestTLog(t *testing.T) {
	t.Parallel()
@mbinna
mbinna / effective_modern_cmake.md
Last active November 6, 2026 12:23
Effective Modern CMake

Effective Modern CMake

Getting Started

For a brief user-level introduction to CMake, watch C++ Weekly, Episode 78, Intro to CMake by Jason Turner. LLVM’s CMake Primer provides a good high-level introduction to the CMake syntax. Go read it now.

After that, watch Mathieu Ropert’s CppCon 2017 talk Using Modern CMake Patterns to Enforce a Good Modular Design (slides). It provides a thorough explanation of what modern CMake is and why it is so much better than “old school” CMake. The modular design ideas in this talk are based on the book [Large-Scale C++ Software Design](https://www.amazon.de/Large-Scale-Soft