Glenn Franxman gfranxman

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.

	Always follow the instructions in plan.md. When I say "go", find the next unmarked test in plan.md, implement the test, then implement only enough code to make that test pass.

	# ROLE AND EXPERTISE

	You are a senior software engineer who follows Kent Beck's Test-Driven Development (TDD) and Tidy First principles. Your purpose is to guide development following these methodologies precisely.

	# CORE DEVELOPMENT PRINCIPLES

	- Always follow the TDD cycle: Red → Green → Refactor
	- Write the simplest failing test first

	#include <opencv2/opencv.hpp>

	#include <pcl/common/common_headers.h>
	#include <pcl/io/pcd_io.h>
	#include <pcl/point_types.h>
	#include <pcl/point_cloud.h>
	#include <pcl/visualization/pcl_visualizer.h>

	#include <Eigen/Core>
	#include <Eigen/LU>

	#!/usr/bin/env python
	"""
	Bundle the site-packages of the current python environment into a zip file

	If you have any wheels at `$(pwd)/wheels`, those will be used instead of
	of the locally installed packages. For instance, if my pwd looked like

	wheels/
	\|-- numpy-1.11.2-cp27-cp27mu-manylinux1_x86_64.whl

	# source this code in a Bash shell
	# and run like django-post-payload '{"a": "b"}'
	# you can set the credentials by exporting DJANGO_USER and DJANGO_PASSWORD

	django-post-payload() {
	local TARGET=http://ws.somesite.com/some/service/endpoint/
	local PAYLOAD=$1
	#"{\"asd\": \"asd\"}"
	local LOGIN=http://ws.somesite.com/login/
	#