Sheru Bhai AI video prompt

One paragraph locks the lion, one directs the shot.

What the prompt has to do

An AI video tool reads text and decides everything else: framing, movement, light, how long the shot lasts. A prompt that only says “Sheru Bhai in an auction” leaves all of that to chance. The prompt on this page sets the two things the tool cannot infer: what the character must look like, and what has to happen inside a very short shot.

Everything is written as direction, not as a wish. The camera move is named, the cut is named, the ending is named. That is the difference between a prompt that produces a usable clip and one that produces a mood board.

The two paragraphs

The first paragraph is the character lock. It describes the mane, the fur, the outfit and the expression, and it ends by telling the tool to carry the description into every later prompt. Keep it identical between clips. If the mane changes between shot one and shot three, the sequence stops reading as one character.

The second paragraph is the shot. It says where he arrives, who moves first, how many steps he takes, where the camera sits, when it cuts and how long the whole thing runs. Change this paragraph freely; change the first one only when you want a different look.

Sample boss entrance prompt

Keep the character

Keep one character locked in: a royal white lion with a thick dark mane, a navy sherwani with gold buttons, thin gold aviators, a single ruby ring and a calm half-smile. Same face, same mane shape, same outfit and the same warm white fur in every shot. Repeat this paragraph in each later prompt so the tool keeps him consistent.

The shot

A vintage black Rolls-Royce rolls to a stop at the palace gates at dusk. Six bodyguards in black reach the door first and hold it open. He steps out, fixes one cuff, and walks forward without hurrying. Camera: low angle, slow motion for three steps, then a hard cut to a close-up of the aviators. Vertical 9:16, 8 to 12 seconds, moody blue hour light.

Timing and frame

Eight to fifteen seconds is the practical window. The sample uses eight to twelve and breaks the action into three beats: the arrival, the walk, the close-up. A vertical 9:16 frame is assumed because the clips are made for phone feeds. If your tool asks for a different ratio, change that line first; the rest of the prompt still holds.

BeatWhat happensWhat the prompt names
Arrival The vehicle stops and the crew moves. Ride, place, who reaches the door first.
Walk He steps out and closes the distance without hurrying. Low angle, slow motion, a fixed number of steps.
Cut A hard cut to the close-up that ends the clip. Aviators, shallow focus, the final frame.

Three beats fit the window. Adding a fourth usually forces the tool to rush one of the first three.

Where to paste it

Paste the full text into the prompt field of whatever tool you use, such as Google Flow or Veo. Tool limits, credits and available lengths change often, so check them on the tool’s own site rather than here. If the field has a character limit, paste the shot paragraph first and the character lock second; the cut is easier to survive than a truncated character.

If the result drifts, do not rewrite everything. Change one line, regenerate, and compare. Camera direction is usually the line that needs adjusting, because “slow dolly-in” and “handheld push” produce very different clips from the same character text.

Copy the two paragraphs, then tune them

Start from the sample, change the ride or the room, and copy the result in one click.

Build this prompt