Two tabs for one decision is the bug.
A builder shipped PolyToro: prediction-market event in, etoro instrument out - one screen.
https://t.co/H1LjceQNpD
illustrative workflow. Capital at risk.
NFA.
Claude Code + a headless browser are f*cking rewriting how product demos get made
A MacBook and an X profile become a fully animated isometric workstation, built entirely through code
No After Effects. No manual keyframes. No screen recording
The video shows a closed laptop opening up, signing in, loading a live profile, lifting off the deck to reveal the board, then settling into a finished desk setup
But the interesting part isn't the animation
It's that the whole scene is a program, not a pile of hand-placed layers
Here's how to recreate it:
MODEL the base, keyboard, hinged lid and desk props as parametric shapes
PROJECT everything into isometric space with one math function
MAP live screen content onto the lid so it follows the hinge angle
ANIMATE each step as a function of time: open, boot, sign in, explode, reassemble
RENDER every frame in headless Chromium and stitch it with ffmpeg
Want to build your own?
Open Claude Code with Playwright and ffmpeg installed and paste:
"Create a premium isometric product demo video using only HTML, SVG and JavaScript
Build a laptop from parametric 3D shapes with a hinged lid, keyboard deck and internal board
Project it into isometric view and map animated screen content onto the display using affine transforms
Animate the sequence: closed device, lid opening, boot, sign-in, profile screen, exploded view, final desk setup
Add a dashboard panel with live status, terminal output, a step progress bar and a spec table
Expose dimensions, timings, colors, text and screen content as editable parameters
Render every frame at 30 fps with Playwright, encode to MP4 with ffmpeg, and give me a reusable script so I can swap the device and rebuild the video"
The real breakthrough isn't one pretty animation
It's one system that can generate a hundred product demos without rebuilding each scene from scratch
I genuinely don't understand why people aren't running this setup
graphify turns any folder into a knowledge graph your agent can search. it's past 128k stars on GitHub and it's in Y Combinator. people aren't just chatting with AI anymore. they're giving it a desk, a memory and an actual job.
hand Claude Code one folder, connect your email, calendar and a browser, and it stops being a chatbot. it reads, it remembers, it does the work, and it double-checks itself before doing anything dumb.
founder, marketer, student, doesn't matter. this is for you.
here's the whole thing:
> make a folder called AGENT, open Claude Code in it, and add a CLAUDE.md with one rule: "save every decision, person and project to memory/"
> run npm install -g @googleworkspace/cli, then gws auth setup and gws auth login. now it can read your Gmail, Calendar and Drive
> run claude mcp add playwright npx @playwright/mcp@latest. now it has its own browser to open sites, fill in forms and grab data for you
> run uv tool install graphifyy, then graphify install, then type /graphify . and the whole folder turns into a knowledge graph it can search in seconds instead of opening files one by one
> run npm i -g jev-guard, then jev-guard key "...", then jev-guard install claude. a tiny model called Jev checks every action before it happens. rm -rf gets blocked, a sketchy email with hidden instructions gets flagged, and it costs about 4 cents per 1,000 checks
> type /schedule and tell it "every morning at 7:30, send me what came in, what you drafted and what's waiting for my yes". now it runs on a routine
15 minutes of setup, and you never open an empty chat again.
all the repos:
-> https://t.co/7MBPMXbhVb
-> https://t.co/nHfHGUY2zQ
-> https://t.co/4jokm3ebW1
-> https://t.co/RfJikkAJX8
bookmark this and test it today