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.

@minniel
minniel / slack_history.py
Last active February 28, 2022 22:02 — forked from Chandler/slack_history.py
Download Slack Channel/PrivateChannel/DirectMessage History
from slacker import Slacker
import json
import argparse
import os
import shutil
import copy
from datetime import datetime
# This script finds all channels, private channels and direct messages
# that your user participates in, downloads the complete history for
@praphull27
praphull27 / Jenkins_Protractor_Headless_Chrome_Setup_Ubuntu_14.04.md
Last active April 9, 2024 11:44
Jenkins, Protractor and Headless Chrome Browser Setup on Ubuntu 14.04

Jenkins, Protractor and Headless Chrome Browser Setup on Ubuntu 14.04

Update Ubuntu

sudo apt-get update
sudo apt-get upgrade

Install Java

@bsmartt13
bsmartt13 / gist:efa02c40ea12c09d9c3a
Created September 17, 2014 17:02
OTX IP Reputation download links (updated hourly)
https://reputation.alienvault.com/reputation.generic.gz
https://reputation.alienvault.com/reputation.generic
https://reputation.alienvault.com/reputation.data
https://reputation.alienvault.com/reputation.snort.gz
https://reputation.alienvault.com/reputation.snort
https://reputation.alienvault.com/reputation.iptables.gz
https://reputation.alienvault.com/reputation.iptables
https://reputation.alienvault.com/reputation.squid.gz
https://reputation.alienvault.com/reputation.squid
https://reputation.alienvault.com/reputation.unix.gz
@igrigorik
igrigorik / json-bq-schema-generator.rb
Last active November 3, 2022 17:34
BigQuery JSON schema generator
require 'open-uri'
require 'zlib'
require 'yajl'
# References
# - https://developers.google.com/bigquery/preparing-data-for-bigquery#dataformats
# - https://developers.google.com/bigquery/docs/data#nested
#
def type(t)
@philips
philips / user-data
Last active September 16, 2020 14:19
Running consul on top of CoreOS utilizing etcd for automatic bootstrapping.
#!/bin/bash
source /etc/environment
name=$(cat /etc/machine-id)
if [ ! -f /opt/consul ]; then
mkdir /opt
mkdir /var/lib/consul
curl http://storage.googleapis.com/ifup-org-tilde/consul-0.2.0-linux-x86-64.lzma | lzcat > /opt/consul
chmod +x /opt/consul
@mreiferson
mreiferson / Dockerfile
Last active April 12, 2016 10:10
nsqd Dockerfile
FROM ubuntu
RUN apt-get -y install wget tar ca-certificates
RUN wget https://s3.amazonaws.com/bitly-downloads/nsq/nsq-0.2.23.linux-amd64.go1.1.2.tar.gz
RUN tar zxvf nsq-0.2.23.linux-amd64.go1.1.2.tar.gz
RUN mkdir -p /usr/local/bin
RUN cp /nsq-0.2.23.linux-amd64.go1.1.2/bin/nsqd /usr/local/bin
RUN mkdir -p /data
VOLUME ["/data"]
@plentz
plentz / nginx.conf
Last active November 8, 2026 01:36
Best nginx configuration for improved security(and performance)
# to generate your dhparam.pem file, run in the terminal
openssl dhparam -out /etc/nginx/ssl/dhparam.pem 2048
@jwalton
jwalton / Dashing EC2.md
Last active October 10, 2019 04:47
EC2 CloudWatch stats for Dashing

Get EC2 CloudWatch stats and graph them in Dashing.