Matthew Sweeney sweenzor

name: tufte-viz description: | Ideate and critique data visualizations using Edward Tufte's principles from "The Visual Display of Quantitative Information." Use this skill when: (1) Designing new data visualizations or charts (2) Critiquing or improving existing visualizations (3) Reviewing dashboards or reports for graphical integrity (4) Deciding between visualization approaches (5) Reducing chartjunk or improving data-ink ratio (6) Planning small multiples or high-density displays

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.

Go to https://script.google.com
Create a New Project
Replace the Code.gs file it creates for you with the javascript below (copy/paste)
Save the script
Go to Triggers (looks like an alarm clock on left-hand side)
Create a Trigger that acts every 10 minutes and calls filterNGPVANSpam
You'll need to authorize this script to act on your behalf, which may require that you use the scary "Advanced" section to allow the script to read/write to your email inbox.

The Zoom install package for macOS is mad. Rather than actually using the installer to install things, it does everything in the preinstall script. That's bonkers, and also means that the system won't have a list of the files it installed, because it's doing it using shell script.

The script appears to install two items, namely:

/Applications/zoom.us.app
~/Library/Internet Plug-Ins/ZoomUsPlugIn.plugin

If the user opening the package isn't an administrator, it looks like it will install the app in the user's home folder instead. If they are an administrator, Zoom will delete the ZoomUsPlugIn.plugin from /Library if it's there, but it still installs to ~/Library.

It also adds Zoom to your Dock automatically, without asking.

	"""
	The most atomic way to train and run inference for a GPT in pure, dependency-free Python.
	This file is the complete algorithm.
	Everything else is just efficiency.

	@karpathy
	"""

	import os # os.path.exists
	import math # math.log, math.exp

	(function () {
	// Define the keyboard shortcut
	const shortcut = {
	key: "N",
	modifiers: ["ctrlKey", "shiftKey"],
	};

	// Function to create a new block, open it in the sidebar, and set focus
	async function createBlockAndFocus() {
	const todayDate = new Date();

	from fastapi import Request, HTTPException
	from pydantic import BaseModel, BaseModel, HttpUrl
	from modal import Secret, App, web_endpoint, Image
	from typing import Optional, List
	from example import proposal
	import os

	app = App(name="circleback", image=Image.debian_slim().pip_install("openai", "pydantic", "fastapi"))

	class Attendee(BaseModel):

	from io import StringIO
	import sys
	from typing import Dict, Optional

	from langchain.agents import load_tools
	from langchain.agents import initialize_agent
	from langchain.agents.tools import Tool
	from langchain.llms import OpenAI

	/* Based on original Dark Age theme by @shodty */

	/* MAIN BODY AND BLOCK COLORS */

	.roam-body-main {
	margin-top: 45px;
	border-radius: 12px;
	background: var(--background);
	margin-right: 6px;
	margin-left: 9px;