You write shot lists for an AI video generator. Turn my idea into an animated short where one character walks through a city with a coffee, and the city keeps happening around her. IDEA: [who walks where, in 1-3 sentences. Example: a woman in a long black coat crosses Manhattan from the river to the crosswalk, coffee in hand] NUMBER OF SHOTS: [default 10, 5 seconds each] Rules for the story: - Follow the walk from start to end, so the shots read as one route, not random scenes. - Invent your own character and outfit. No famous studio characters. - Each shot has one small city moment next to her: a street painter, a gust from a passing train, a hot dog stand, a lost tourist with a map, rain on the pavement. - She stays calm and keeps walking. The city never stops for long, and neither does she. - Mix wide street shots, side tracking shots and close-ups of her face. - Keep her in every shot, and describe her with the same words every time. - The last shot closes the walk with the best moment. - No real people, no logos, no on-screen text, no violence. Output format: a numbered list. Each shot is one self-contained English prompt, ready to paste, because the generator doesn't remember earlier shots. Each prompt must contain, in this order: 1. Who is in the frame and what they do, with one clear action 2. The setting: the street, the station, the weather 3. The camera: one continuous shot, with its angle and movement 4. This style line, word for word: "Stylized animated film with a hand-painted look, soft hazy daylight, muted colors with one red accent, clean expressive character design. No text, no subtitles, no logos, no watermark." 5. Format: 16:9, 5 seconds After the list, add one line with the order of the shots and a suggestion for music.
This clip was generated with Kling 3 by Kuaishou. The prompt on this page is the exact text used to create it.
The example was generated with Kling 3 using: 135:76 · 0:45 · Animation · Cinematic · Cityscape. Reusing these settings gives the closest match.
Copy the prompt as-is first, then change the subject while keeping the camera, motion, and lighting terms. Video models are sensitive to wording, so expect a few runs.
Yes. Prompts work best as starting points — swap scenes, adjust pacing terms, or change the aspect ratio to fit your project.