Skip to content
templatebeginnerFree

Cinematic scene prompt template

A fill-in-the-blanks structure for text-to-video prompts that consistently produces composed, cinematic shots instead of generic stock footage.

You'll need: Access to a text-to-video model

Fill in your details

Values are inserted into every prompt below. Nothing you type leaves your browser.

9 field(s) left — unfilled variables stay as {{placeholders}}.

  1. 1Fill the template

    Fill every slot, even with a simple choice. Empty slots are filled by the model's defaults, which is why outputs look generic. Order matters: most models weight earlier words more heavily, so put the shot and subject first.

    {{camera}}, {{lens}} look. {{subject}} {{action}} in {{setting}} at {{time_of_day}}. {{lighting}}. Mood: {{mood}}. Shallow depth of field, natural motion, film grain, cinematic colour grade. Duration {{duration}}.
    Avoid: text, watermarks, distorted faces or hands, sudden cuts.
  2. 2Know what each slot does

    Camera: shot size + angle + one movement, e.g. 'Low-angle medium shot, slow dolly-in'. Lens: '24mm wide' exaggerates space; '85mm' compresses and isolates the subject. Subject: who or what, with 2-3 visual details. Action: one verb phrase, present tense. Setting: specific place with one texture detail. Time of day and lighting: these do most of the work for mood ('golden hour backlight', 'overcast soft light', 'neon practicals'). Mood: one or two words, not a list.

  3. 3Iterate one slot at a time

    Change only one slot between generations so you learn what caused the difference. Keep a log of prompt, seed (if available) and result.

Example input

Fisherman coiling rope on a boat at dawn, quiet determination, slow dolly-in, 50mm anamorphic.

Example output

Low-angle medium shot, slow dolly-in, 50mm anamorphic look. An elderly fisherman in a yellow oilskin jacket coils a wet rope onto the deck in a small wooden boat in a grey harbour at dawn. Soft overcast light with a warm glow on the horizon. Mood: quiet determination. Shallow depth of field, natural motion, film grain, cinematic colour grade. Duration 6 seconds. Avoid: text, watermarks, distorted faces or hands, sudden cuts.

Troubleshooting

  • The camera ignores your movement: move the camera slot to the very start and use common terms (dolly-in, pan left, orbit).
  • The result looks like stock footage: make the subject and setting more specific and add one imperfect detail (rust, wet surfaces, wind).
  • Some models ignore 'Avoid' lines: use the tool's negative-prompt field if it has one, otherwise rephrase positively ('clean frame with no text').

Known limitations

  • Lens and film terms are interpreted loosely; they steer the look rather than simulate optics.
  • Supported durations differ by model; the duration slot may be ignored and set in the interface instead.

More from Video generation