Alex Shemetovskiy AlexKVal

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.

name	capture-skill
description	Capture learnings, patterns, or workflows from the current conversation into a new or existing skill. Use when the user wants to save what was learned, discovered, or built during a conversation as a reusable skill for future sessions.

Capture Skill from Conversation

This skill helps you extract knowledge, patterns, and workflows from the current conversation and persist them as a reusable skill.

When to Use

Delete unused or obsolete files when your changes make them irrelevant (refactors, feature removals, etc.), and revert files only when the change is yours or explicitly requested. If a git operation leaves you unsure about other agents' in-flight work, stop and coordinate instead of deleting.
Before attempting to delete a file to resolve a local type/lint failure, stop and ask the user. Other agents are often editing adjacent files; deleting their work to silence an error is never acceptable without explicit approval.
NEVER edit .env or any environment variable files—only the user may change them.
Coordinate with other agents before removing their in-progress edits—don't revert or delete work you didn't author unless everyone agrees.
Moving/renaming and restoring files is allowed.
ABSOLUTELY NEVER run destructive git operations (e.g., git reset --hard, rm, git checkout/git restore to an older commit) unless the user gives an explicit, written instruction in this conversation. Treat t

Use cases

Optional/nullable data types.
Variant data types.
Flag with extra information needed only in some cases (or different in different cases).
Modeling states in a state machine where each state can have some data associated with it.
List of operations for batch processing, where each operation can have its own arguments.
Alternative to function overloading that allows avoiding combinatorial explosion when there are multiple parameters that need variants.
Nested data structures with heterogenous nodes.

	function Identity(v) {
	return { val: v };
	}

	function chain(m,fn) {
	return fn(m.val);
	}

	// ^^^ that's it!

	ruby '2.7.1'

	gem 'rails', github: 'rails/rails'
	gem 'tzinfo-data', '>= 1.2016.7' # Don't rely on OSX/Linux timezone data

	# Action Text
	gem 'actiontext', github: 'basecamp/actiontext', ref: 'okra'
	gem 'okra', github: 'basecamp/okra'

	# Drivers

	#!/bin/bash

	############################################################################################
	# WARNING: THESE SCRIPTS AND CHANGES ARE NOT MEANT FOR ANYBODY BUT ME.
	# RUNNING THESE WILL MESS UP YOUR COMPUTER IN MYSTERIOUS AND USUALLY UNRECOVERABLE WAYS.
	############################################################################################

	if [[ "$(sw_vers -productVersion)" != 10\.15* ]]; then
	echo "This is only meant to run on macOS 10.15.* Catalina" >&2
	exit 1

	Pre reqs:

	I installed these using homebrew

	* nodejs
	* yarn

	Create an application directory manually

	$ mkdir app_name

	##### Windows
	# Windows thumbnail cache files
	Thumbs.db
	Thumbs.db:encryptable
	ehthumbs.db
	ehthumbs_vista.db

	# Dump file
	*.stackdump