Fernando da Silva schoulten

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.

Conventional Commit Messages

See how a minor change to your commit message style can make a difference.

git commit -m"<type>(<optional scope>): <description>" \
  -m"<optional body>" \
  -m"<optional footer>"

"Help, I keep getting a 'Permission Denied (publickey)' error when I push!"

This means, on your local machine, you haven't made any SSH keys. Not to worry. Here's how to fix:

Open git bash (Use the Windows search. To find it, type "git bash") or the Mac Terminal. Pro Tip: You can use any *nix based command prompt (but not the default Windows Command Prompt!)
Type cd ~/.ssh. This will take you to the root directory for Git (Likely C:\Users\[YOUR-USER-NAME]\.ssh\ on Windows)
Within the .ssh folder, there should be these two files: id_rsa and id_rsa.pub. These are the files that tell your computer how to communicate with GitHub, BitBucket, or any other Git based service. Type ls to see a directory listing. If those two files don't show up, proceed to the next step. NOTE: Your SSH keys must be named id_rsa and id_rsa.pub in order for Git, GitHub, and BitBucket to recognize them by default.
To create the SSH keys, type ssh-keygen -t rsa -C "your_email@example.com". Th

	FROM python:3.10-slim

	ENV APP_HOME /app
	WORKDIR $APP_HOME

	# Allow statements and log messages to immediately appear in the Knative logs
	ENV PYTHONUNBUFFERED True
	ENV PORT 8080

	RUN pip install --no-cache-dir --upgrade shiny

	# install.packages("devtools")
	# devtools::install_github("tchiluanda/rsiconfi")

	#Criar gráfico de sankey
	df_desp <- get_dca(2018,"I-D","1")



	df_desp_trabalho<-
	df_desp %>%

	# install.packages("devtools")
	devtools::install_github("tchiluanda/rsiconfi")

	library(rsiconfi)
	library(dplyr)
	library(tidyr)

	###########Trabalhando as despesas

	#Monta um vetor com todos os códigos de UFs do Brasil

	#' Procura a palavra ou frase do segundo vetor que melhor
	#' se aproxima do primeiro. Particularmente útil para
	#' corrigir nomes de municípios.
	#'
	#' @param x Vetor de strings de referência.
	#' @param y Vetor de strings a serem buscados.
	#'
	#' @return vetor com as strings de y próximos
	#' de x.
	#' @export

	# Get a person's name, location, summary, # of connections, and skills & endorsements from LinkedIn

	# URL of the LinkedIn page
	user_url <- "https://www.linkedin.com/in/daattali"

	# since the information isn't available without being logged in, the web
	# scraper needs to log in. Provide your LinkedIn user/pw here (this isn't stored
	# anywhere as you can see, it's just used to log in during the scrape session)
	username <- "yourusername"
	password <- "yourpassword"