Nikola Balic nibzard

Extracting & Deduping Slide Screenshots From a YouTube Talk

A playbook + lessons learned from pulling a Keynote deck out of a long YouTube video. The slides lived in a small picture-in-picture box in the top-right corner. Goal: get one image per distinct slide, then OCR the deck to Markdown. We only cared about a 15-minute window (05:05:00 → 05:20:00).

Pipeline at a glance

Stealth_Bench_V1 — Run Analysis (browser-use + Steel + claude-opus-4-8)

Run: Stealth_Bench_V1 · framework browser-use (installed) · browser steel · model claude-opus-4-8 Started: 2026-06-19 18:39:20 (local) · Duration: ~6.4 h (23,066 s) · Tasks: 80 completed Headline: 52 pass / 28 fail (65%). ~93% of failures are bot-detection, not infra.

Run data: full per-task traces and the summary JSON are published at https://github.com/steel-experiments/bu-bench-results.git — see run_data/Stealth_Bench_V1_framework_browser-use_browser_steel_model_claude-opus-4-8_start_at_20260619_183920/ (80 task .json traces) and results/Stealth_Bench_V1_framework_browser-use_browser_steel_model_claude-opus-4-8.json.

RFC: Changelog and Release Conventions

Status

Proposed

Goal

Adopt a release workflow that:

Atlas SDK Specification

Deep-research framework where the primary builder is a coding agent, not a human developer.

Status: Design proposal. Not yet implemented. Current @steel-dev/atlas (v0.1.x) ships a single opinionated research() function and CLI. This document specifies how to evolve it into an SDK for building domain-specific researchers, with Agent Experience (AX) as the primary design constraint.

0. Why this exists

Steel × ClawBench Collaboration Plan

Why This Makes Sense

ClawBench runs 153 tasks on live production websites (Uber Eats, Booking.com, Robinhood, etc.) inside local Docker containers with a DIY Chromium stealth layer. Steel solves exactly this problem as a managed service. The integration surface is minimal — browser-use already accepts a custom CDP endpoint, meaning the harness change is one line of code.

ClawBench is also the most credible AI browser agent benchmark published to date (arXiv

RFC: Skill Overlays

Status: Draft Last Updated: 2026-03-09 Author: @nibzard

Abstract

Skill overlays are an optional extension for specializing a base skill without forking it.

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.

Hacker News [dead] Post Analysis

Date: April 6, 2026
Method: Steel browser session (logged in as nibzard with "showdead" enabled), JS-based DOM extraction
Scope: 23 pages of /newest (~690 stories), 100 domains investigated via /from?site=

What is a [dead] post on HN?

Posts marked [dead] have been killed by HN moderators or the anti-spam algorithm. They are invisible to logged-out users and to users who haven't enabled "showdead" in their profile. They are NOT the same as HTTP 404s — the linked URLs may still work fine. It's a content moderation signal.

The Evolution of Steel CLI: From Node.js to agent-browser Integration to Rust

Prologue: The Original Vision

The story of the Steel CLI begins in April 2025, when the first lines of code were committed to create a tool for Steel.dev — a browser automation platform. The initial commit on April 20, 2025 (52ea53c) established a foundation built with Pastel, a React-based CLI framework that allowed developers to create beautiful terminal interfaces using React components.

feat: init pastel cli tool, added api and login authentication

 <!DOCTYPE html>
 <html lang="en">
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
 <title>Context Rot — A Cross-Disciplinary Research Review</title>
 <style>
   :root {
     --bg: #fafafa;
     --text: #1a1a2e;