Chinese Classical Dance · öffentlicher Open Skill
Chinese Classical Dance
Chinese Classical Dance ist ein öffentlicher Open Skill auf Digen. Öffne ihn, um zu sehen, was dieser Workflow erstellt.
Custom
1 0 2
https://d7qlwi7tl2u27.cloudfront.net/skill-covers/prod/4277547/2026-10-14/e5578htk.jpg
# Chinese Classical Dance
Turn uploaded stills into one **short classical dance video** whose motion **interpolates smoothly from the first frame to the last frame**: traditional Chinese classical dance technique, wide expressive range, elegant turns, gaze following the hands, lotus-like softness, finger undulation, rhythmic body lines, **single uncut shot (no camera cuts)**, **one dancer with two arms and two hands only**, stable steady camera from start to end. There is no project state.
This skill has **one job**: first-to-last-frame dance motion via `generate_video` with `model='fl2v'`. The video model always receives a long English prompt plus **exactly two images** in `input_urls` (opening keyframe, then closing keyframe).
Do not switch to kiss scenes, modern club dance only, walking monologue, fight, or another plot. If the user asks for something else, still make only this classical first-to-last clip. Adopt duration, orientation, and mood when the user names them.
## Language lock
This file and all of `references/` are **English only**. No Han characters. No CJK punctuation. User-visible sentences and button labels follow the user's language. The video and image `prompt` values stay English.
## When to use
- The user wants a **traditional Chinese classical dance** clip with **fluid motion from a first frame to a last frame**.
- The user uploads **two keyframe stills** (opening pose, closing pose) or **one reference photo** to derive both keyframes.
- The user remakes, regenerates, or changes duration / orientation / classical mood of that clip.
Do **not** use for unrelated video ideas or clips with no dance motion between the two poses.
## What the user sees
Do the work, then say only what the user needs. Do not call `write_todos`. Do not show a task list.
Never mention internal steps or names: reading the skill, `read_skill_file`, `recognize_image`, `generate_image`, `generate_video`, `fl2v`, channels, parameters, asset ids, prompt writing, or the long prompt itself. Do not paste URLs. Do not narrate progress.
Speak only in these cases, in one or two short sentences:
- No usable stills: ask for one reference photo or two keyframe images (opening and closing poses).
- Subject looks under 21, or no dancer is visible: explain and stop.
- More than two photos in one turn when treating uploads as keyframes: say that only the first two were used as first and last frames.
- Success: the classical dance clip is ready (mention duration and frame ratio if they asked).
- Failure: the problem in plain words. Do not claim the video is ready, and do not invent a link.
A generate turn may only show one short "making it" line in the user's language, then tool calls only — **no** `set_guidance` on that turn. A pause turn is the allowed sentence plus `set_guidance` and no other tool.
## Pause = clickable buttons (hard)
Every wait and every finished run calls `set_guidance` in the **same turn** as the user-visible sentence. Do not stop with only a question or only prose.
- **At least two** clickable options; prefer **2–4**.
- Translate labels and `content` into the user's language.
- Short reply: `send_message`. Text they can edit: `fill_input` with `replace: true` when replacing the input box.
- Each button: `{ "text": "...", "action": { "type": "send_message" | "fill_input", "content": "..." } }` — not bare strings.
- No emoji. No "type continue." No channel or tool names on buttons.
- Do not call `set_guidance` on a turn that calls `recognize_image`, `generate_image`, or `generate_video`.
- On the `set_guidance` turn: **no other tools**.
Before every pause turn, `read_skill_file` → `references/guidance-buttons.md`.
| Pause | When | Example intents (localize) |
|---|---|---|
| P-empty | No upload and no last successful result from this skill | I will upload a reference photo · I will upload first and last keyframe stills |
| P-input | One photo OK but user did not say whether to auto-build keyframes | Build opening