Fetching from the wire…
Vibe Coding2026-09-25 · source-backed
LaunchVideo (293 HN points) takes a URL or product description, Opus 5.5 writes the film as code, and a serverless agent renders it deterministically at 1080p with Playwright, Chromium and ffmpeg in microVMs. The page puts it at roughly 90K input and 15K output tokens per video. (launchvideo.io) The full agent, renderer and web app are in diggerhq/shipvideo, created September 24, 88 stars. Very early. "Deterministic render from code the agent wrote" is the part I'd steal for any generative-media pipeline, because it makes the output diffable.
Each link below shares sources, entities, or timing with this story.
Every browser-automation setup I've built in the past year has the same shape. A Playwright process. An MCP wrapper around it. A headless Chromium that drifts from the browser I actually use, with a different cookie jar, a different rendering path, and a login session I have t...
Claude Code, Gemini CLI, Codex, Cursor, TaskMaster, Junie, and OpenCode, plus a web UI and headless Chromium with Playwright (GitHub). Useful if you actually want to benchmark agents head-to-head without managing seven installs. I'd use it for exactly that, then run my daily w...
alibaba/open-code-review cut v1.12.1 at 11:23 UTC this morning. It's a Go CLI that ran as Alibaba's internal review assistant for two years before the May 2026 open-source release, and it's at 24,596 stars with 443 added today. The architecture explains the claimed numbers. It...
The maintainer shipped a full modernization on August 15: Node-only, ES6, TypeScript definitions, Playwright and Puppeteer support, node:test, Biome, and the removal of CasperJS and PhantomJS, with Opus 4.8 writing most of it (stephen-cresswell.com). The argument beyond the re...
1. Flip your multi-model pipeline to review-then-generate. Instead of using a reasoning model to plan before code generation, let the specialist generate freely and use reasoning tokens for review. Paper shows 90.2% pass@1 vs 87.2% for the planning pattern. Source 2. Audit you...
A solo Claude Opus 4.5 agent spent $9 and 20 minutes building a retro game. It was broken. The same model, wrapped in Anthropic's multi-agent harness, spent $200 over 6 hours and produced a fully playable game with physics, sprite editors, and AI integration. Anthropic's engin...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.