Someone open-sourced our own video pipeline.

The exact clicks

1. Install it with one line in a terminal (a box where you type one instruction and press enter): npx hyperframes init.
2. Describe what you want, in plain language, to an AI coding agent (like Claude Code) — the same way you'd describe it to a person.
3. The agent writes a "recipe" for you: plain text where each clip is one step, with a start time and how long it lasts.
4. Run one more command — npm run render — and it turns that recipe into a real video file.
5. Open the finished file yourself to confirm it's real, the same thing we did before publishing this.

What changes

• The same idea this channel has run by hand for fifteen episodes — instruction in, real video out — is now a free, open-source tool anyone can install.
• It's free (Apache 2.0 license — no cost, no usage fees) and it installs its own starter files for AI coding agents, so an agent like Claude Code already knows how to use it.
• We measured it ourselves: a 10-second test video rendered in about 15 seconds on ordinary hardware.

What it will not do

• This is a tool for building videos with code/agent instructions — it doesn't replace a camera or footage of a real event; it's for the same kind of screen-recording-and-graphics video this channel already makes.
• We tested a small, simple example, not a full multi-scene production — a longer, more complex video will take longer to render and may need more setup than shown here.
• Open-source projects change; if a command in this episode no longer matches what you see, check the project's own current documentation rather than assuming this episode is still exact.