和風ファンタジー · সর্বজনীন Open Skill

和風ファンタジー

鳥居の向こうは、別の国。 和風ファンタジーの世界を、実景＋一つの不可能で作ります。 ・8拍構成 — 鳥居、階、狐、灯、月、社、光、余韻 ・一カットに不可能はひとつだけ — ファンタジー小物を並べると安っぽくなります ・光 — 黄昏の青＋暖色の魔法光源をひとつ。光には必ず発生源を見せます ・音 — 鈴ひとつ、笛の一節、風、低いドローン。古い音であって壮大な音ではありません ・和風の意匠 — 千本鳥居、石段、白狐、提灯、御札、彼岸花 画面内の文字はすべて日本語です。既存作品の世界観・意匠は使用しません。オリジナル制作のみです。

@omegabubble

Anime

0 1 3

https://d7qlwi7tl2u27.cloudfront.net/skill-covers/prod/4267104/2026-10-18/w5whsr9x.png

---

allowed-tools:
  - generate_image
  - generate_video
  - generate_music
  - mix_clips
  - recognize_video
  - download_video
---

# 和風ファンタジー / Japanese Fantasy World

Japanese fantasy is **a real landscape with one impossible thing**: a torii with too many gates, a fox with two tails, a moon too large.

## When to use

- The user asks for 和風ファンタジー, 異世界, 幻想, or a Japanese fantasy world.
- The user wants an anime-style fantasy establishing shot or world reveal.
- The user wants a vertical fantasy short for TikTok / Shorts / Reels.
- The user uploads a fantasy concept art and wants it animated in place.

Do **not** use this skill for: existing franchise worlds or copyrighted designs — see Notes.

## Steps

### 0. Clarify — 和風ファンタジー

Call `write_todos` with the plan: clarify → (optional) reference analysis → choose the setting → storyboard → stills → shots → sound → edit.

Ask only what you still need:

1. **Which setting?** — 場所を2〜3個から選ぶ
2. **Which hour?** — 朝 / 昼 / 夕方 / 夜
3. **With people?** — なし / 後ろ姿だけ
4. **Frame?** — portrait 9:16 / landscape 16:9

Read the choice back in one Japanese line and get a yes.

### 1. Reference analysis (only if the user supplies a clip or image)

```
recognize_video(video_url=<url>, prompt='Analyse this as a Japanese 和風ファンタジー video. Report: the light and hour, average shot length, how static the camera is, how much of the frame is a close-up, whether people are shown face-on, and the sound design. Do not describe the plot.')
```

For a YouTube / TikTok / Instagram link, resolve it first with `download_video(url=...)`.
Turn the answer into a short treatment note and read it back. Then build an **original** piece.

### 2. Storyboard — 8 beats, about 30 seconds

| # | Beat | Length | What happens |
|---|---|---|---|
| 1 | 鳥居 | 4s | A torii in fog, then another behind it |
| 2 | 階 | 4s | Stone steps rising into cloud |
| 3 | 狐 | 4s | A two-tailed fox, seen once, distant |
| 4 | 灯 | 4s | Lanterns floating without support |
| 5 | 月 | 5s | An oversized moon over a ridge |
| 6 | 社 | 4s | A shrine half-swallowed by forest |
| 7 | 光 | 5s | Light spilling from a doorway in the rock |
| 8 | 余韻 | 4s | Wide: the impossible landscape. Hold |

**Rule: exactly one impossible element per shot — never a crowd of fantasy props**

### 3. 日本語での作法 — what makes it Japanese

**All on-screen text is Japanese. No English anywhere.**

**日本の符号（必ずどれかを入れる）**
- 鳥居、千本鳥居、石段、社、祠
- 狐、白狐、面、提灯、御札
- 月、霧、雲海、竹林、彼岸花
- 灯籠、人魂、光の道
- 和風の意匠：青海波、麻の葉、市松
- 音：鈴、笛、太鼓、風

**光**
Dusk blue with one warm magical source. The impossible light must have a visible origin.

**音**
A single bell, a flute phrase, wind, low drone. Fantasy should sound old, not epic.

**テロップ**
- One line only: `鳥居の向こうは、別の国。` — mincho, white, small

### 4. Stills

```
generate_image(model='t2i', prompt='Japanese fantasy landscape: a long corridor of vermilion torii gates disappearing into fog, stone steps rising into cloud, dusk blue light, one warm lantern glow, no people, no text, painterly anime background art', name='和風ファンタジー', entity_type='storyboard_images', orientation='landscape')
```

```
generate_image(model='i2i', input_urls=[<user asset>], prompt='the same scene from a closer angle, more texture, no people, no text', name='和風ファンタジー 2', entity_type='storyboard_images')
```

### 5. Shots

```
generate_video(model='i2v', input_urls=[<still asset>], prompt='almost completely static: subtle ambient movement only, a very slow camera push, no people, no camera shake', name='和風ファンタジー', entity_type='storyboard_videos', orientation='landscape')
```

Keep `orientation` identical across every shot; never write a physical model name — channels only.

### 6. Sound

```
generate_music(model='music', prompt='quiet Japanese ambience for 和風ファンタジー, ambient textures, no drums, no melody, 45 seconds', name='和風ファンタジー', entity_type='music')
```

A single Japanese line through `tts.design` is optional. **Confirm the exact Japanese with the user first** (kana + romaji) — a wrong reading r