Wesley C. Hinkle evokateur

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.

🧠 How to Save Context Tokens When Using Claude Code

This is a personal reference workflow for minimizing token usage while maintaining project continuity across Claude Code (Sonnet 4 with file access).

✅ Setup: Populate `CLAUDE.md`

Claude loads CLAUDE.md automatically at session start.

⚙️ ADHD Meds Master Guide (Ritalin & Stimulants)

🔑 Core Insight

Both caffeine and ADHD meds (e.g., Ritalin) boost focus, not clarity or wisdom. Without clear priorities, extra stimulation can lead to efficient distraction rather than productivity.

🧭 Guiding Principle

本文由简悦 SimpRead 转码，原文地址 www.anthropic.com

A blog post covering tips and tricks that have proven effective for using Claude Code across various ......

Engineering at Anthropic

We recently released Claude Code, a command line tool for agentic coding. Developed as a research project, Claude Code gives Anthropic engineers and researchers a more native way to integrate Claude into their coding workflows.

Claude Code is intentionally low-level and unopinionated, providing close to raw model access without forcing specific workflows. This design philosophy creates a flexible, customizable, scriptable, and safe power tool. While powerful, this flexibility presents a learning curve for engineers new to agentic coding tools—at least until they develop their own best practices.

	# Apple Silicon laptops with firmware > 13.0 have a native charge threshold that does not required any userspace daemon running.
	# This native limit works even when the laptop is sleeping or powered off therefore it is preferable to the userspace daemon.
	# Nonetheless, it only works with fixed thresholds (80% as upper limit and 70% as lower limit).
	# CHWA key is the one used to enable/disable the native limit. 01 = 80% limit, 00 = no limit
	##
	typeset -g smc_command="/usr/local/bin/smc"
	typeset -g smc_charge_limit_key="CHWA"
	typeset -g smc_charge_limit_status_on="01"
	typeset -g smc_charge_limit_status_off="00"

	# include following in .bashrc / .bash_profile / .zshrc
	# usage
	# $ mkvenv myvirtualenv # creates venv under ~/.virtualenvs/
	# $ venv myvirtualenv # activates venv
	# $ deactivate # deactivates venv
	# $ rmvenv myvirtualenv # removes venv

	export VENV_HOME="$HOME/.virtualenvs"
	[[ -d $VENV_HOME ]] \|\| mkdir $VENV_HOME

	# Ignore everything in the root except the "wp-content" directory.
	/*
	!.gitignore
	!wp-content/

	# Ignore everything in the "wp-content" directory, except the "mu-plugins", "plugins", and "themes" directories.
	wp-content/*
	!wp-content/mu-plugins/
	!wp-content/plugins/
	!wp-content/themes/

	*~
	.DS_Store
	.svn
	.cvs
	*.bak
	*.swp
	Thumbs.db

	# wordpress specific
	wp-config.php

	# vim:fileencoding=utf-8:ft=conf:foldmethod=marker

	#: Fonts {{{

	#: kitty has very powerful font management. You can configure
	#: individual font faces and even specify special fonts for particular
	#: characters.

	font_family JetBrains Mono Medium
	bold_font JetBrains Mono Bold

	# -- mode: ruby --
	# vi: set ft=ruby :

	PASSWORD_PATH = ".password"
	PASSWORD_ID_PATH = ".password_id"

	# Make sure to have installed vagrant-triggers plugin
	# > vagrant plugin install vagrant-triggers

	# After the first `vagrant up` stop the VM and execute the following steps