Toc Toc Video Prompts: How the Fixed Effect Is Written (and What to Change)

Most AI video tools hand you a prompt box and wish you luck. The toc toc generator does the opposite: the prompt is written, locked and shipped as the product, so the dance, the crew and the camera look the same on every run. This guide opens up what that prompt says — the action, the camera, the costume and scene language, and the guardrails — and explains what you can and cannot change.

The short version

• Fixed effect. You do not write a prompt; the prompt is part of the product and is locked.
• What it describes: the toc toc moves, a crew of costumed friends, home lighting, a steady eye-level camera, a party mood.
• Guardrails: keep the subject recognizable (same face, same hair, same look), no text overlays.
• Your variables: the photo you upload, and how many videos you generate.
• Output: an 8-second vertical (9:16) MP4, 30 credits per video.
• Try it: make a toc toc video.

Why the prompt is fixed

A dance trend lives or dies on consistency. If every run produced a different dance, a different crew and a different camera, the result would feel random and the trend would stop being recognizable. Fixing the prompt means every video is unmistakably the same effect, in the same way a CapCut template always produces the same edit.

That design choice also removes the hardest part of prompt writing. You do not need to know how to describe a hop, a turn or a costume — the description already exists, and it has been written to keep motion readable at eight seconds.

What the prompt specifies

Here is what the fixed effect asks the model for, grouped the way it is actually written.

Action (the dance). Rhythmic one-foot hops, one fist knocking on an invisible door in time with a playful beat, quick quarter-turns, arms flung wide between hops. The subject copies the same hops and knocking moves and clearly has fun doing it.

Characters (the crew). A small group of quirky costumed friends joins the subject: a flamboyant showman in a burgundy velvet robe with a gold chain, a dancer in a leopard-print suit, and a tall friend in a pinstripe suit.

Camera and motion. A vertical phone video, steady eye-level camera, bright home lighting, comedically high-energy party mood, and a slight motion blur on the hops so the movement reads as motion rather than as a freeze-frame.

Identity guardrail. Keep the subject recognizable — same face, same hair, same look as the photo. This is the instruction that ties the output back to your input.

Delivery constraints. No text overlays, and light party audio rather than any copyrighted track.

Prompt layer	What it pins down	Fixed or yours?
Action	Hops, knocking fist, quarter-turns	Fixed
Characters	The costumed friend crew	Fixed
Camera and light	Vertical, steady eye-level, bright home lighting	Fixed
Mood and motion	Party energy, slight motion blur on hops	Fixed
Subject identity	Same face, hair and look as the photo	Fixed (depends on your photo)
Source photo	Who is in the video, how clearly they read	Yours

Step by step

Step 1 — Understand the split. The prompt is not yours to edit; the photo is. Read the table above and decide what you actually control.

Step 2 — Feed the guardrail a good photo. The identity instruction works best on a sharp, well-lit, front-facing photo with one subject. This is the single highest-leverage “prompt” change available to you.

Step 3 — Generate and inspect. Check the subject’s identity and whether all three moves read: the hops, the knocking fist and the turns.

Step 4 — Retry or re-shoot. If a detail is off, re-run once — each generation is a fresh take — and if it still drifts, change the photo.

Writing your own extension is not a thing here

Two constraints are worth stating plainly, because they trip up people arriving from general video tools:

• There is no prompt field to edit. The effect ships locked, so a “better prompt” cannot be entered. If you have seen dance-video prompts circulating for other tools — describing the move, the camera, the beat — that language is already inside this effect and is not exposed for editing.
• Two natural-language prompts are not the same as one effect. In a free-form generator, adding “and a crew of friends in costumes” to a prompt competes with every other instruction you wrote. Here that crew is described once, carefully, and applied on every run.

If you want to compare this locked-effect approach against the more manual photo-selection workflow, the photo-to-dance guide covers the input side in detail. And if you want the moves themselves explained, start with the trend guide.

What actually changes a result

Since the prompt is fixed, your levers are small but real:

• Photo sharpness. A crisp source keeps identity stable across the dance.
• Framing. Full body or knees-up gives the moves room; a tight headshot shows less.
• Subject count. One subject removes ambiguity; a crowd creates it.
• Retries. A second run on the same photo samples a different take, which often resolves a weak result.

None of these require writing a line of prompt text — which is exactly the intended trade. For the arithmetic of how retries interact with credit packs, see the cost guide.

A useful mental model: treat the source photo as the only field in a form you are allowed to fill in. Everything the effect does — the moves, the crew, the camera, the mood — is a fixed template, the way a CapCut template fixes an edit. You choose the subject and the number of takes; the effect chooses everything else.

Use the effect as designed

The value of a locked prompt is that it removes the guesswork. Choose the photo, generate, keep the take that reads best. Open the toc toc generator to try it, and check pricing so you know what each run costs before you start.

Frequently asked questions

Can I write my own prompt for the toc toc video?

No. The prompt is fixed and ships as part of the effect — there is no prompt field to edit. Your inputs are the photo and the number of runs.

What does the fixed prompt describe?

The dance (one-foot hops, a knocking fist, quick quarter-turns), a crew of costumed friends, vertical framing on a steady eye-level camera with bright home lighting, a party mood with slight motion blur, and a guardrail that keeps the subject’s face, hair and look consistent with the photo.

Can I change the crew, the costumes or the scene?

No. Characters, costumes, camera and lighting are part of the locked effect. Changing the photo is how you change who is dancing; everything else stays the same by design.

Can I upload a reference video to control the motion?

No. The product takes a single photo. There is no reference-video input.

Is there a way to make the result more consistent with my photo?

Yes — use a sharp, well-lit, front-facing, single-subject photo, framed from the knees up if possible. The identity guardrail in the prompt is only as strong as the input it is given.

What do one of these videos cost?

30 credits per video, with one-time packs starting at $9.90 and no subscription. See pricing for the pack breakdown.