SuperAPI AI Directory

seedance-travel-city-walk

Design & creation

View repository

Updated 10-16 10:18

About this skill

Cinematic travel vlog montage · Cinematic travel roaming

One traveller moves through a place scene by scene, each scene carrying its own timecode, its own location and one short line. The polish is bought with film grain and golden-hour light.

This Skill carries one prompt structure distilled from 31 human-verified Seedance cases on goodcase.ai. It is a sibling of seedance-prompt-library (all templates in one Skill); install this one when you only want this kind of clip. Do not invent structure from general video-generation knowledge: follow the structure below and ground the draft in one anchor case from references/cases.md.

Use when

  • Destination diaries, city walks, hikes, camping trips and departure sequences where one traveller has to stay the same person across six or eight locations.
  • Chinese: For films like destination diaries, city roaming, hiking, camping, and setting off on a journey: one traveler must remain the same person across six to eight locations.

Workflow

  1. Collect inputs. Ask only for what is missing: subject or product, setting, duration, aspect ratio, whether reference images / audio exist, and which language the final prompt should be in.
  2. Pick one anchor case. Read references/cases.md, choose the case closest to the request, and name it in the reply. One anchor, never an average of several.
  3. Fill the structure block by block. Every block below must be present in the final prompt; an empty block is where prompts go vague.
    1. Opening line: runtime, aspect ratio, the format named as a cinematic travel vlog, and who the traveller is
    2. Look block written as parameters: film stock, grain, colour grading, depth of field, frame rate, handheld feel
    3. One-sentence consistency lock covering hair, makeup, outfit and expression across every scene
    4. Scene blocks by timecode, each headed with a location name: arrival, a landscape wide, an activity beat, a food or slow-living beat
    5. Spoken lines written inside the scene where they are said, one short sentence each
    6. Closing scene at golden hour or at night, ending with her looking into the lens and signing off
    7. Tail: voice and lip-sync requirements, then the exclusions for text, logo and watermark
  4. Apply the guidance and check the pitfalls (next two sections) before returning anything.
  5. Return the finished prompt as one copy-pasteable block, the anchor case link, and a three-line checklist of what to verify in the generated video.

Guidance (from the cases)

  • Head every scene with both a timecode and a place name. The Bali diary writes Scene 3 (8-12s) — Rice Terrace & Jungle Moments, and each block holds one location, one action and one camera move.
  • Put the cinematic feel in a parameter list, not in adjectives. The Bali diary states 4K cinematic video, 24fps, 35mm film grain, realistic handheld camera and adds warm vintage grading.
  • Keep the lines short and tied to what just happened. Bali uses Don't film this part — actually, keep filming it. after she wobbles on the board, and closes on Goodnight from Bali.
  • Lock the gear as well as the person. The Korean camping case writes Maintain the same woman, outfit, SUV, tent, campsite, and equipment throughout, so the car and tent cannot redesign themselves halfway.
  • When the trip involves real physical work, spell out the mechanics. The camping case asks for realistic tent fabric, flexible poles, stakes and bans instant tent setup.

Pitfalls

  • Packing eight scenes into thirty seconds and giving every one of them a line. The Bali diary runs eight scenes and only four of them speak.
  • Naming the location and leaving the camera to the model. Each scene needs its own move, written like Camera trails her from behind, then swings into a close-up.
  • Mixing in cheap phone-footage words such as shaky phone video or low quality. The handheld here sits on top of film grain and shallow depth of field, and image quality has to stay up.
  • Turning the film into a string of empty landscape shots. Give every scene a concrete action: Bali has her walking barefoot at the tideline and drinking coconut water through a paper straw.

Language

Reply in the user's language. The prompt itself can stay in English when that is what the user's Seedance workflow expects; ask if unclear. The Chinese version of this structure follows.

Chinese: Cinematic travel roaming

One person walks through a place scene by scene, each scene with its own timecode, its own location, and one short line of dialogue. The texture is carried by film grain and golden hour light; the cheap phone flaws approach does not apply here.

Structure:

  1. Opening line: write the duration, aspect ratio, and genre as cinematic travel vlog, and who this traveler is
  2. Write the texture section as a parameter list: film stock, grain, color grading, depth of field, frame rate, handheld feel
  3. A one-line consistency lock governing hairstyle, makeup, clothing, and expression, covering every scene
  4. Break into scenes by timecode, each with a location name: arrival, a big scenic landscape shot, an activity, a scene of eating or slowing down
  5. Write the dialogue into the scene where it is spoken, one short line per scene
  6. Place the ending at golden hour or at night, and finally she looks at the camera and says goodbye
  7. Ending: voice and lip-sync requirements, then exclude on-screen text, logos, and watermarks

Key points:

  • Each scene's title gives both the timecode and the location. The Bali one is written as Scene 3 (8-12s) — Rice Terrace & Jungle Moments, one paragraph covers just one location, one action, one camera movement.
  • Write the cinematic feel as a parameter list, don't pile up adjectives. The Bali one directly writes 4K cinematic video, 24fps, 35mm film grain, realistic handheld camera, then adds warm vintage grading.
  • Keep dialogue short, and stick it to what just happened. In Bali, after she loses her balance on the board, she says Don't film this part — actually, keep filming it., and the whole film ends on Goodnight from Bali.
  • Lock the gear together with the person. The Korea camping one writes Maintain the same woman, outfit, SUV, tent, campsite, and equipment throughout, so the car and tent won't change appearance midway.
  • When the journey has real physical labor, write out the mechanical process. The camping one requires realistic tent fabric, flexible poles, stakes, and bans instant tent setup。

Common pitfalls:

  • Cramming eight scenes into thirty seconds, and having every scene speak. In the Bali one, only four of the eight scenes have dialogue.
  • Only writing where they went, leaving the camera movement to the model. Every scene must give its own camera action, written as Camera trails her from behind, then swings into a close-up like this.
  • Mixing in cheap phone flaw words, like shaky phone video, low quality. The handheld here is built on film grain and shallow depth of field, so the image quality must go up.
  • Shooting the whole thing as a string of empty landscape shots with no people. Give each scene a specific action; the Bali one is walking barefoot across the tide line and drinking coconut water with a paper straw.

Copy-ready lead-in / Copy-ready lead-in

For users who would rather paste into a chat model than run this Skill, hand them this line above the structure:

I want a cinematic travel vlog montage. [The destination is Kyoto in November, at the peak of the autumn leaves.] [On camera: a young woman with a backpack, walking from an early-morning alley to the riverbank at dusk.] Using the prompt template below, rewrite it into one ready-to-use Seedance video prompt for me:

I want to make a cinematic travel roaming video, [the destination is Kyoto, the time is the November red leaf season], [the person on camera is a girl with a backpack, walking from the early morning alleys all the way to the riverside in the evening]. Please use the prompt template below to help me rewrite it into a Seedance video prompt that can be used directly:

Notes

  • references/cases.md is generated from data/ in LearnPrompt/awesome-seedance and refreshed daily; do not hand-edit an installed copy.
  • Prompts in the reference file belong to their creators (linked on every card). Transfer the structure; do not present a close copy as original work.
  • For the full library across every model, install the goodcase Skill from LearnPrompt/goodcase-lite.