IGOR KRAVTCOV kravik

REPO HERE: https://github.com/tandpfun/wardrobe

name: extract-clothing-cutouts description: Extract high-quality, deduplicated transparent ecommerce clothing cutouts from a folder of photographs where people wear one or more garments. Use when Codex must find outfit or model photos, identify unique clothing across images, create focused references, reconstruct complete garments with Imagegen, remove a solid chroma background into RGBA PNGs, and output only the finished clothing images into a new folder under the current working directory.

Extract Clothing Cutouts

Turn photographs of worn clothing into source-faithful standalone catalog PNGs. Treat each result as a reconstruction from visible evidence, not literal segmentation whenever the wearer or another layer occludes part of the garment.

name

codex-plan-mode

description

Используй, когда нужно сначала подготовить полный по решениям план реализации отдельным PLAN-файлом, без правок реализации до завершения плана. Навык исследует репозиторий без изменений, задаёт значимые вопросы через request_user_input, создаёт PLAN-NN файл в docs/03_plans и затем спрашивает, переходить ли к реализации.

Файловое Планирование

Этот навык нужен, когда пользователь просит сначала спланировать работу и сохранить результат отдельным файлом. Не выполняй реализацию, пока план не создан и пользователь не выбрал переход к реализации в финальном вопросе.

Работай в 3 фазах и через диалог доведи план до состояния, в котором его можно сразу передать другому инженеру или агенту. Хороший план должен быть подробным по намерению и реализации, а также полным по решениям: исполнитель не должен принимать недостающие решения.

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.

OpenClaw Prompts - Build Your Own AI Assistant

Prompts to recreate each piece of the OpenClaw system. Use these with any AI coding assistant.

1. Personal CRM "Build a personal CRM that automatically scans my Gmail and Google Calendar to discover contacts from the past year. Store them in a SQLite database with vector embeddings so I can query in natural language ('who do I know at NVIDIA?' or 'who haven't I talked to in a while?'). Auto-filter noise senders like marketing emails and newsletters. Build profiles for each contact with their company, role, how I know them, and our interaction history. Add relationship health scores that flag stale relationships, follow-up reminders I can create, snooze, or mark done, and duplicate contact detection with merge suggestions. Link relevant documents from Box to contacts so when I look up a person, I also see related docs."

2. Meeting Action Items (Fathom)

Reverse Engineering SP110e LED Controller

Through Bluetooth sniffing, using this method I've been able to (semi reliably) control the SP110e controller via Homebridge.

You can find this controller for very cheap here: https://www.aliexpress.com/item/4000773623427.html?spm=a2g0o.productlist.0.0.4f09329cJ7C1H4&algo_pvid=542e757b-587f-4540-8652-2195883f1349&algo_expid=542e757b-587f-4540-8652-2195883f1349-0&btsid=0bb0622a16012309671478585ed4bd&ws_ab_test=searchweb0_0,searchweb201602_,searchweb201603_

This entire guide is for Bluetooth LE libraries, and uses hexidecimal.

	//------------------------------------------------------------------------
	// The SwiftUI Lab: Advanced SwiftUI Animations
	// https://swiftui-lab.com/swiftui-animations-part1 (Animating Paths)
	// https://swiftui-lab.com/swiftui-animations-part2 (GeometryEffect)
	// https://swiftui-lab.com/swiftui-animations-part3 (AnimatableModifier)
	//------------------------------------------------------------------------
	import SwiftUI

	struct ContentView: View {

	//
	// ViewController.swift
	// PanOnScroll
	//
	// Created by Shinichiro Oba on 2018/03/08.
	// Copyright © 2018 bricklife.com. All rights reserved.
	//

	import UIKit

	import Foundation
	import UIKit
	import CoreML
	import Vision

	// Quick and dirty example of how to fix Vision VNPixelBufferObservation BAD_ACCESS.
	// Please handle errors in swift by yourself, this code sample intended only for demonstration.
	// MLModel must have multiArray output and Image input.
	// Before using this functions install CoreMLHelpers from https://github.com/hollance/CoreMLHelpers

	import Vision
	//1
	let sourceImage = UIImage(named: "jony.jpg")
	var resultImage = sourceImage
	//2
	let detectFaceRequest = VNDetectFaceLandmarksRequest { (request, error) in
	//4
	if let results = request.results as? [VNFaceObservation] {
	//5
	for faceObservation in results {

	//
	// GameViewController.h
	// pulseeffectobjectivec
	//
	// Created by Lachlan Hurst on 10/3/17.
	// Copyright © 2017 Lachlan Hurst. All rights reserved.
	//

	#import <UIKit/UIKit.h>
	#import <SceneKit/SceneKit.h>