ruphy

// SPDX-License-Identifier: MIT
//
// This code will add a GPT3() function in your Google Sheets
// This code is originally inspired from https://twitter.com/fabianstelzer/status/1572926883179778050
// To use it, insert your API key below, open Google Sheets -> Extensions -> Apps Script -> Copy & Paste this -> Save
//
// Usage: =GPT3(prompt, max_tokens (default=15), model (default=davinci))
// Example usage: =GPT3("Once upon a time,", 1000, "davinci")

var API_KEY = "your-API-key";

semanticentity

javascript:(async function(){
  if(!navigator.mediaDevices||!navigator.mediaDevices.getDisplayMedia){
    alert("Screen recording not supported");
    return;
  }

  var filename = prompt("Make sure you are in a tab other than the one you are recording.\nEnter filename (no extension):","lewm-recording");
  if(!filename) filename = "lewm-recording";

  alert("Select a TAB and check 'Share tab audio' in the popup. For audio, do NOT chose full screen or window.");

remorses

interface InputFile {
    path: string
    start?: number // Start time for trimming (in seconds)
    end?: number // End time for trimming (in seconds)
    fadeIn?: number // Fade in duration in seconds
    fadeOut?: number // Fade out duration in seconds
}

interface ConcatenateOptions {

nikivdev

My 2Do tasks

This is just contents of my never ending lists of tasks I tagged in 2Do with read, watch and check tags.

All lists are sorted by priority. I hope to bring all these lists closer to 0 when I get time. But until then, perhaps you will find something interesting in them.

Here are the lists I share so far.

• Read
• Watch

deepakness

Tool / approach	Cost (typical)	Ease of setup	Maintenance burden	Scalability	Zero-downtime support	Laravel-specific features	Backups	Monitoring/logs	SQLite friendliness	Hetzner fit
Forge	Paid: $12–$39/mo

championswimmer

How Personal AI Agents and Agent Orchestrators like OpenClaw or GasTown are Made

Over the last few months, projects like Gas Town by Steve Yegge and OpenClaw by Peter Steinberger have made “AI agent orchestrators” feel suddenly mainstream. It is tempting to treat them as a new kind of intelligence, but under the hood they are still a small set of primitives wired together with discipline: an LLM API call, a state loop, tools, memory, and orchestration.

This raises a practical question: what is actually inside an “agent,” and how is it different from ChatGPT (a chat UI over a model) or coding tools like Claude Code (an agentic coding surface)? Gas Town’s README frames it as a “multi‑agent orchest