Yossi Elkrief MaTriXy

OKF Graph Wiki — Agent-Maintained Knowledge Graph for Context Retrieval

I've been running Andrej Karpathy's LLM Wiki pattern for several months — reading sources, compiling them into a compounding, git-diffable wiki instead of re-deriving everything from scratch every session — and I'm a genuine convert. But two things kept nagging at me: query salience (finding the right page reliably, not just something plausible) and token economy (not re-reading half the wiki to answer one question).

Xplain Privacy Notes

Xplain is designed as an opt-in, local-first browser assistant.

Xplain does not sell user data, does not use user data for advertising, does not use analytics, and does not require an account.

What Xplain Reads

Xplain reads page content only after the user presses Read current page. It can capture:

name: zte-dev-loop description: > Zero-Touch-Engineering development loop — an MM-NATIVE gate-pipeline layer. It rides on top of the /mm PM skill (which owns the HANDOFF/Section-0 lifecycle, inbox protocol, insights, closing ritual, and fresh/continuing/update modes) and adds: a gated, cross-family-reviewed pipeline (brainstorm -> TDD implement -> audit -> security -> tests -> visual QA -> works-locally -> cross-family judge -> land) that drives a feature from idea to landed code fully autonomously except the one human deploy gate. This is a user-approved ZTE operating MODE that overrides a few mm opt-in defaults for one task (see Part Z). Builders run via the autonomous /yolo session (spawned per mm Section 4.6.1); audits/judges run cross-family (e.g.

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	regression-dog
description	Review code changes for regressions.

Review code changes for behavioral differences between the before and after code.

Important:

Do NOT run tests, typechecks, linters, or build commands. CI already handles those. Focus your context budget entirely on reasoning about the code changes and its implications on logic/behavior/data/etc.

Build a pixel-perfect Space Invaders arcade game in a single self-contained HTML file with inline CSS and JS. Replicate the original 1978 Taito arcade cabinet experience with full fidelity to the original rules, sprite designs, and mechanics.

CANVAS & DISPLAY

800×600px canvas (or scaled 2× from original 224×256 logical resolution)
Black background, phosphor-green or white pixel aesthetic

	# ══════════════════════════════════════════════════════════════════════
	# SETUP — install everything this file depends on
	# ══════════════════════════════════════════════════════════════════════
	# Core tools (one line). Run once on a fresh machine:
	#
	# brew install starship fzf zoxide eza bat
	#
	# starship — the prompt
	# fzf — fuzzy finder (Ctrl+R history, Ctrl+T files)
	# zoxide — smart cd (`z <partial>` jumps to frequent dirs)

	---
	name: viral-content-factory
	description: Create viral social media content (TikTok/Instagram carousels and outfit-change videos) for app promotion. Generates AI characters, carousel image sets via Gemini, and animated mannequin videos via Kling 3/fal.ai. This skill should be used when the user wants to create new carousel content, mannequin videos, or expand the content library for any character.
	tools: Read, Write, Edit, Bash, Glob, Grep, Agent
	---

	# Viral Content Factory

	End-to-end system for generating viral TikTok/Instagram content promoting a product through AI-generated characters. Creates two content types:
	1. Carousels — sets of 7-8 styled images per topic (Gemini image generation)

	# Sharing Claude Code Context Files Across a Team (Without Building Anything)

	## The Problem

	You're a small team using Claude Code. Everyone keeps `.md` context files. Some are shared, some are private, and some should only be visible to part of the team. Git alone doesn't handle per-file access control, and Google Drive isn't dev-friendly. You need something in between.

	## The Solution: GitHub Org + Teams + Submodules

	No middleware. No new service. Just GitHub features you already have.

	#!/bin/bash
	#
	# find-axios-repos.sh
	# Scans all repos in a GitHub org for axios dependency usage.
	# Searches package.json files for axios in dependencies and devDependencies.
	#
	# Requirements: gh CLI (authenticated)
	#
	# Usage: ./find-axios-repos.sh <org-name>