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.

GitHub Pull Requests Search Results

Date: 2024-11-01

Search Query: is:pr is:open label:docs

Title URL Created At Labels
gh-126180: Remove getopt and optparse deprecation notices Link 2024-10-31T05:35:32Z docs; awaiting core review; DO-NOT-MERGE
Explain what a DBM database is and point to shelve earlier Link 2024-10-31T02:32:13Z docs; skip issue; awaiting review; skip news
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
@amcasari
amcasari / amcasari-tech-management-bits.md
Last active January 11, 2021 14:42
amcasari's tech management theories, lemmas, and corollaries

my core values

  • Enable a growth mindset
  • A team is more than a collection of individuals
  • Sustainability is always a priority
  • Give cover for calculated risks
  • Build soapboxes, shine spotlights

on working with me (and space i’ll definitely give you)

  • I work spiky :) This is sometimes due to personal commitments, community organization work, talks, coffees, my brain getting interrupted by things that are not
@PhiSYS
PhiSYS / multiple-gource.sh
Last active February 19, 2026 19:58
Create multiple repositories visualization video using gource and ffmpeg
#!/usr/bin/bash
declare -a repos=(
"$HOME/Projects/myproject/testing/.git"
"$HOME/Projects/myproject/testing-bundle/.git"
"$HOME/Projects/myproject/service-skeleton/.git"
"$HOME/Projects/myproject/documentation-web/.git"
"$HOME/Projects/myproject/documentation-api/.git"
"$HOME/Projects/myproject/coding-standard/.git"
"$HOME/Projects/myproject/context/microservice1/.git"
@betatim
betatim / example.ipynb
Created July 2, 2020 07:47
Notebook that contains various kinds of elements
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
@mislav
mislav / gh-rename-master
Last active April 24, 2022 10:02
Rename the default branch of a repository using GitHub CLI https://github.com/cli/cli/releases/tag/v0.10.0
#!/bin/bash
# Usage: gh-rename-master <newbranch> [<remote>]
#
# Renames the "master" branch of the current repository both locally and on GitHub.
#
# dependencies: GitHub CLI v0.10
set -e
newbranch="${1?}"
remote="${2:-origin}"
@betatim
betatim / does-it-limit.py
Created March 30, 2020 20:56
Highly scientific rate limit checker
import requests
import time
S = requests.Session()
now = time.time()
while True:
r = S.get(YOUR_HOST_HERE)
now_ = time.time()
delta_t = now_ - now
@skrawcz
skrawcz / PyBayFormatter.py
Last active October 8, 2021 22:46
JSON Structured Logger
import json
import logging
import sys
logger = logging.getLogger(__name__)
class PyBayFormatter(logging.Formatter):
"""Implementation of JSON structured logging that works for most handlers."""
@1st1
1st1 / example.py
Last active April 18, 2026 20:08
asyncio queues example
import asyncio
import random
import time
async def worker(name, queue):
while True:
# Get a "work item" out of the queue.
sleep_for = await queue.get()