wowkin2

Збудуй мені Telegram-бота — персонального коуча по цілях, який щодня пушить власника і веде облік прогресу в Google Таблиці.

Стек (не обговорюється)

• Cloudflare Worker, plain JS, деплой npx wrangler deploy. Ніяких залежностей, крім @anthropic-ai/sdk.
• База — Google Sheets через сервісний акаунт (JWT RS256 → oauth2.googleapis.com, скоуп spreadsheets).
• Мозок — Claude API (модель claude-opus-5, effort low) з tool use.
• Розклад — cron triggers воркера. Все крутиться в хмарі 24/7, нічого не залежить від ноутбука.

Що я дам (спитай і постав секретами через wrangler secret put)

• TELEGRAM_BOT_TOKEN — токен від BotFather

wowkin2

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.

wowkin2

Browser Tools

Minimal CDP tools for collaborative site exploration.

Start Chrome

./browser-tools-start.js              # Fresh profile
./browser-tools-start.js --profile    # Copy your profile (cookies, logins)

wowkin2

How to clean Docker.raw image

Many times I stuck in situation when Docker consumes whole 70Gb, then I remove all Containers and Images in Docker, and only Volumes remaining which should consume ~7Gb. But storage is not free anymore. Hard to say what is using it. It can probably be just reserved by and not released. Found the way to backup volumes, drop virtual machine, create new and restore volumes.

Steps

• To create backups - create file from example below and run that:

chmod +x backup_volumes.sh
./backup_volumes.sh

wowkin2

#!/bin/bash

# Before running make file executable:
# chmod +x decompile.sh

# And check Android SDK version (in this case it is - 33.0.1)

# Create directory
mkdir "decompiled"

wowkin2

import rospy

from std_srvs.srv import Empty


def reset_world():
    """
    Reset Gazebo world to restore initial position
    """
    rospy.wait_for_service('/gazebo/reset_world')

wowkin2

$('.right').each(function(a, b){
    var value = $(this).html();
    if (value !== undefined) {
        var red = 255 - parseInt(value.slice(0, -1)) * 2;
        var green = 0 + parseInt(value.slice(0, -1)) * 2;
        var color = 'rgb(' + red + ', ' + green + ', 0)';
        $(this).css('background', color);

        
        console.log(value, color);

wowkin2

import os


CRLF = b'\r\n'
LF = b'\n'

allowed_extensions = [
    'gitignore', 'Procfile', 'Docker',
    'txt', 'md', 'ini', 'json', 'yaml',  # NOTE: do not add csv, they can be huge
    'js', 'css', 'scss', 'html', 'htm', 'svg',

wowkin2

import ssl
import socket
import asyncio
import logging
import collections
import time

from asyncio import sslproto
from email._header_value_parser import get_addr_spec, get_angle_addr
from email.errors import HeaderParseError

wowkin2

/**
Following script will parse all photos on imageshack.com
and will prepare list of files with appropriate names (like it they were originally)
to be downloaded on the next step

Scroll till the end of the page, so all images will be lazy-loaded.
Press F12 in browser copy-paste following code and press Enter.
Copy result and add it to the following Python script.
**/
var data = [];