Skip to main content
One sentence is enough: the agent picks the look, the pacing and the structure for you. Say more only about what must come out exactly as you imagine it. This page shows how to say it, and the topic pages show the words for motion, sound, text, color and brand.

​
One-time setup

In the Studio app there is nothing to set up. With your own agent, install the skills once: one paste. Then start each request by calling the skill: /hyperframes in Claude Code and Cursor, $hyperframes in Codex, or however your agent invokes a skill.

​
Describe it, or hand it something

Describe the video when the idea is in your head: length, shape, mood, and the key elements.
Create a 10-second product intro with a title over a dark background and quiet music.
Hand it something when you have material: a URL, a doc, a spreadsheet, a transcript. The agent reads it and writes about something real, which gives richer videos than invented copy.
Turn this CSV into an animated bar chart race.
A full example, a 45-second launch built from a live site:
Make a 45-second 1920x1080 launch video for https://linear.app. Energetic but minimal, use the site’s own palette and screenshots. Structure: hook stating the problem, 3 feature beats with UI captures and one-line captions, end card with logo + “Try it free”. Female TTS voice, confident tone, subtle electronic BGM under -18dB.
Made from the prompt above. At the other end, a dense visual spec rebuilt a broadcast-style globe in one go (how): One-shot render from the distilled spec, no iteration.

​
Before the build

The agent asks only about what it can’t decide well, then builds. Want to approve the plan first? Ask for a storyboard (in the Studio app, turn on Start with a storyboard): you see the video as stills before anything moves. In a hurry, say Just build it, don’t ask me anything.

​
The shape of a strong prompt

Each part of this skeleton removes a decision agents most often get wrong on a first try:
  • Route with a slash command, only with your own agent: /hyperframes loads the framework rules (/slideshow for a deck). The Studio app does this for you.
  • Spec duration and resolution up front. Defaults are 1920x1080 at 30fps.
  • Beats get timestamps. Include pacing instructions (“then hold on the blinking cursor”) — agents skip breathing room unless told.
  • Copy is quoted exactly, with / for line breaks. Unquoted copy gets paraphrased.
  • Technique: name registry blocks exactly as they appear in the catalog — they’re adapted starting points, not drop-ins (blocks ship with demo content the agent rewrites to your beat, so naming one pins the technique). Pin a technique wherever the default choice can fail (see the specification dial).
  • Negatives close the gaps: “no narration” is not “silent” — if you want no sound at all, say “no audio”. Avoid ambiguous phrases like “no external assets” (CDN-loaded runtimes are normal; say “no image or media files” if that’s what you mean).
Assembled:
Make an 8-second 1920x1080 video. Beat 1 (0-4s): dark macOS terminal types “npx skills add heygen-com/hyperframes” character by character, then hold on the blinking cursor. Beat 2 (4-5s): the terminal shatters into fragments. Beat 3 (5-8s): bold white kinetic text on black slams in word by word, snappy: “YOU JUST MADE THIS / WITH HYPERFRAMES.” Adapt the code-typing and vfx-shatter registry blocks; hand-author the kinetic text. No narration, no image or media files.
Made from the prompt above.

​
Inside each beat

The skeleton above structures the ask. Inside each beat, describe the content with the same five slots every time:
One sentence per element, slots in any order: “a giant number (element) counts up to $4.2M with an odometer roll (motion), easing out as it lands (timing), centered (layout) in green on dark navy (style).” Elements you don’t describe, the agent designs — which is fine when you trust its taste and a drift risk when you don’t (see the specification dial).

Practice it: The brief

Free 20-minute lesson with a short video, a quiz and a brief to try. A brief is the goal, who it is for, two to four specifics and the exact words in quotes.

​
How much to specify

How much you specify sets how far the result can drift from what you imagined, not whether it works. Mood words get a designer’s interpretation; exact colors, timings and named techniques get your frame. Three settings:
  1. Mood words — “a warm, minimal logo sting”. Fastest to write; the agent owns the look.
  2. Style tokens — exact hexes, named textures (“halftone-dot”, “paper-cut matte”, “8-bit pixel”), type direction (“condensed caps”, “thin geometric sans, wide tracking”). Cheap precision, no downside.
  3. Full visual spec — every element gets shape + color + position + timing (see visual specs). Use when you’re recreating a specific look.
Density never hurts — a stronger model just uses less of its own taste. It matters most for technique choices a smaller model won’t default well. Two worth pinning every time:
  • 3D: any real depth, lighting, or camera motion → say “Three.js via the adapter”. CSS perspective transforms read flat on lighting-critical scenes, and Three.js is a supported seek-safe runtime.
  • Sequencing: resolve simultaneity in the text. “The counter fades out and READY stamps in at 4s” produces an overlap; “fades out fully by 4.2s; at 4.2s READY stamps in” doesn’t.

​
Iterate toward the look

After the first version, talk to the agent like a video editor instead of starting over:
Make the title 2x bigger.
Swap to dark mode.
Add a fade-out at the end and a lower third at 0:03 with my name and title.
The captions are too small and they overlap the lower third. Move them up and shrink them.
Replace the background music with assets/track.mp3.
The agent already has the composition open and the skills loaded — small targeted edits produce better results than long re-specifications.

​
Make each change count

Three moves keep iteration convergent instead of circular:
  1. One variable per edit. Change one thing, render, judge, repeat. A request that moves three things at once (“bigger title, warmer colors, faster cuts”) makes it impossible to attribute what helped and what hurt.
  2. Layer one element type at a time. Building up a complex scene? Start with the minimal version (subject + its motion only), confirm it reads, then add exactly one layer per pass — camera move, then background motion, then style treatment. Every addition stays attributable.
  3. Strip, then re-layer. When a scene keeps misfiring, don’t pile on corrections — strip it back to the simplest version that works (freeze the camera, simplify the motion, clear the background), confirm, then reintroduce complexity one layer at a time until you find the ingredient that breaks it.

​
Calibrate with absolute targets

When you’re dialing in a look, relative corrections oscillate — “make the dots finer” overshoots, “a bit bigger” overshoots back. State targets as absolute values and the agent lands them in one pass:
  • Avoid: make the dots 2x finer
  • Instead: dot radius = 25% of row spacing, with clear gaps between dots
  • Avoid: the glow is too strong
  • Instead: glow at ~35% peak opacity, always subtler than the orb itself
And freeze what already works: “the framing and motion are right — don’t touch them; only change the dot size.” Without the freeze clause, a rebuild can drift on axes you’d already settled.

Practice it: Review and correct

Free 20-minute lesson with a short video, a quiz and a brief to try. Watch the first video all the way through, with sound, before you reply.

​
Rules and common mistakes

Ask for a hold, but say what stays alive. A landed title should hold still; keep moving only what is actually changing (a live number, footage, the camera travelling to the next thing). Don’t ask for “breathing” to fill time. Give every second a job. Ask for 12 seconds with 5 seconds of content and the agent has to invent the rest. Either shorten the length or say what the extra seconds show. Write the timing out. “At 4s the counter fades out and READY stamps in” overlaps for a few frames; “the counter is gone by 4.2s; at 4.2s READY stamps in” doesn’t. Quote on-screen text. Unquoted copy gets paraphrased; tagline: "Ship faster." renders word for word. Don’t ask for React / Vue components. Compositions are plain HTML with data-* attributes and a GSAP timeline; framework components force a translation step.
  • Avoid: build a React component for the intro
  • Instead: build the intro scene (the agent writes composition HTML directly)
Don’t over-spec resolution or framerate. Defaults (1920×1080, 30fps) render fast and look great; higher specs slow rendering meaningfully.
  • Avoid: render in 4K 60fps (for a social clip)
  • Instead: say nothing — or 4K only when the delivery target actually needs it
With your own agent, don’t skip the slash command. Without /hyperframes, the agent guesses at HTML video conventions instead of loading the framework’s actual rules.
  • Avoid: make me a video of...
  • Instead: /hyperframes make me a video of...
Don’t paste raw error logs. check localizes the problem first (lint, then a browser gate); a bare log makes the agent re-derive what the tool already knows.
  • Avoid: pasting 200 lines of console output
  • Instead: check reports a missing asset in scene 2 — fix it
Don’t assume the agent knows your assets. It will look, but a path skips the search.
  • Avoid: use my logo
  • Instead: use assets/logo.svg
Don’t hard-time a verbatim script. With supplied narration text, duration follows the spoken words.
  • Avoid: a 60-second explainer from this script: ...
  • Instead: a ~60-second explainer from this script: ...
The agent’s own rules (what it checks before rendering, the lint rules) live in the /hyperframes skill; you don’t need them to ask for a video.