seedance（Seedance 视频提示）

Seedance — Seedance 2.0 Motion Graphics Skill

Generate cinematic motion graphics videos using ByteDance's Seedance 2.0 model via Fal AI. Turn app screenshots into glossy liquid glass promo videos controlled from Claude Code.

What This Skill Does

• Generate image-to-video animations from a single image (with optional end frame control)
• Generate reference-to-video motion graphics from multiple app screenshots
• Generate text-to-video motion graphics from pure prompts
• Scrape App Store screenshots automatically for any iOS app
• Upload reference images to get public URLs for the API
• Apply proven liquid glass prompting patterns that produce Apple-keynote-quality output

Setup

1. Get a Fal AI API Key

Sign up at fal.ai and create an API key from your dashboard.

2. Store the Key

Add to your .env file:

FAL_KEY=your-fal-ai-key-here

3. File Upload (Included)

Seedance needs publicly accessible URLs for reference images. Fal AI includes its own CDN upload — no extra accounts or keys needed. Your FAL_KEY handles both generation and file upload.

Want a full Generations tab with gallery, styles panel, and multi-model support (Kling, Veo, GPT Image, Nano Banana)? Check out the Rubric dashboard at robonuggets.com.

Endpoints

Seedance 2.0 has 5 endpoints on Fal AI. Default to Pro — it's only ~25% more expensive but noticeably better quality.

When to Use Which

• Image-to-video — you have a single image and want to animate it with motion. Supports start + end frame control for precise transitions. Best for: animating a hero image, creating A→B transitions between two states, lip-sync animations.
• Reference-to-video — you have multiple screenshots, logos, or visual assets to blend into a motion graphics piece. This is the primary mode for promo videos.
• Text-to-video — no visual references available (e.g. GitHub repos, docs pages, abstract concepts). Pure prompt-driven.

Parameters

• Total files across all modalities must not exceed 12
• Reference images in prompts: @Image1, @Image2,... @Image9
• Reference videos: @Video1, @Video2, @Video3
• Reference audio: @Audio1, @Audio2, @Audio3
• Max resolution is 720p — there is no 1080p option

Cost Reference (Pro, 720p)

Fast tier is ~20% cheaper. Use for quick tests before committing to Pro.

Sweet Spot (Tested)

These are the proven ratios from extensive testing:

The golden rule: 1 reference image per 2 seconds of video.

• Fewer refs = model focuses better on each one
• More refs (6-7+) = model blends too loosely, loses clarity
• Always describe each image's role in the prompt

Prompting Guide

Rule 1: Logo References First (CRITICAL)

NEVER describe logos or brand elements in the prompt text. Seedance mangles text rendering — wrong shapes, wrong colors, hallucinated details.

Before writing ANY prompt:

1. Ask for or find the app/brand logo image
2. Save it locally
3. Upload it to get a public URL
4. Pass it as one of the @Image references

If you don't have the logo: STOP and ask. Never proceed without it.

Recommended approach: Generate the motion graphics without text/logos in the video at all, then overlay the logo in post-production (Premiere, After Effects, etc). This produces the cleanest results.

Rule 2: The Liquid Glass Prompt Template

This is the proven prompt structure that produces Apple-keynote-quality motion graphics:

Design a motion-graphics style ad using glossy liquid glass design
language. Pure black background. @Image1 is [describe what it shows].
@Image2 is [describe]. @Image3 is [describe]. The [element] from
Image 1 becomes translucent glass [describe transformation].
[Element] from Image 2 [describe]. [Continue for each image].
Multiple camera angles: close-up on glass reflections, pull back
to reveal the full scene. No text, no logos, no words. Style: liquid
glass morphism, Apple Vision Pro aesthetic, premium 3D depth,
self-luminous forms on absolute black, [accent color] accent lighting.

Key rules:

• Describe each @Image by number and what it contains
• Describe how each element transforms into glass
• End with a style line including the brand's accent color
• "Multiple camera angles" keeps it dynamic
• "No text, no logos, no words" prevents garbled text
• Keep under ~200 words — the model ignores long prompts

Rule 3: Device Framing

For product shots, specify the device in the prompt:

Floating desktop screens (best results):

Show the desktop screens floating in 3D space on a pure black
background, tilted at slight angles like a MacBook product shot.
The UI elements on screen become translucent glass with reflections
and refractions.

iPad:

An iPad Pro floating in empty black space, tilted at a cinematic
angle like an Apple product shot. The iPad has visible black bezels
and a physical frame — only the screen content has the glass effect.

MacBook:

A MacBook Pro floating in empty black space, open at a cinematic
angle. The MacBook screen displays the dashboard. Light catches
the aluminium edges.

Floating in black void produces better results than placing devices in environments (offices, desks). Keep it pure black.

Rule 4: Audio

• generate_audio: true adds ambient SFX and music
• May trigger content policy violations on abstract visuals — if it fails, retry with generate_audio: false
• Audio costs the same whether enabled or not
• When it works, the generated audio adds real production value

Rule 5: Negative Guidance

Add constraints to prevent common issues:

• "No text, no logos, no words" — prevents garbled text
• "No cuts" — keeps it as one continuous shot
• "No camera shake" — keeps it stable
• "Pure visual motion only" — reinforces no text

The Full Pipeline

Step 1: Get Screenshots

Option A — Screenshot a website:

from playwright.sync_api import sync_playwright

with sync_playwright() as p:
    browser = p.chromium.launch(headless=True)
    page = browser.new_page(viewport={"width": 1920, "height": 1080})
    page.goto("https://example.com", wait_until="networkidle", timeout=30000)
    page.wait_for_timeout(2000)
    page.screenshot(path="screenshot.png")
    browser.close()

Option B — Scrape App Store screenshots:

import json, urllib.request

# Step 1: Find the app ID
search_url = "https://itunes.apple.com/search?term=APP_NAME&entity=software&limit=3"
data = json.loads(urllib.request.urlopen(search_url).read())
app_url = data["results"][0]["trackViewUrl"]

# Step 2: Scrape screenshot URLs from the page using Playwright
# Look for <source> elements with srcset containing "mzstatic.com"
# The URLs use format like /460x996bb.webp — swap to /1290x2796bb.jpg for full res

Option C — Manual screenshots — take them yourself, save locally.

Step 2: Upload for Public URL

Seedance needs publicly accessible URLs for reference images. Use Fal AI's built-in CDN upload (default) — no extra accounts needed.

Fal AI CDN (default — uses your existing FAL_KEY):

# Step 1: Initiate upload
INIT=$(curl -s -X POST "https://rest.alpha.fal.ai/storage/upload/initiate" \
  -H "Authorization: Key $FAL_KEY" \
  -H "Content-Type: application/json" \
  -d '{"file_name":"my-screenshot.png","content_type":"image/png"}')

# Extract URLs from response
FILE_URL=$(echo "$INIT" | python -c "import sys,json; print(json.load(sys.stdin)['file_url'])")
UPLOAD_URL=$(echo "$INIT" | python -c "import sys,json; print(json.load(sys.stdin)['upload_url'])")

# Step 2: Upload the file
curl -s -X PUT "$UPLOAD_URL" \
  -H "Content-Type: image/png" \
  --data-binary @/path/to/screenshot.png

Use FILE_URL as the public URL in your Seedance API calls. Hosted on v3b.fal.media with long-lived caching. Included with your Fal AI account — no extra cost.

Alternative: Kie AI (if you have a key):

curl -s -X POST "https://kieai.redpandaai.co/api/file-stream-upload" \
  -H "Authorization: Bearer $KIE_API_KEY" \
  -F "file=@/path/to/screenshot.png" \
  -F "uploadPath=generations" \
  -F "fileName=my-screenshot.png"

Returns downloadUrl. Files expire after 3 days.

Alternative: Litterbox (no account needed):

curl -s -F "reqtype=fileupload" -F "time=1h" \
  -F "fileToUpload=@/path/to/screenshot.png" \
  https://litterbox.catbox.moe/resources/internals/api.php

Returns a direct URL. Expires after 1 hour.

Step 3: Generate the Video

Always use a temp JSON file to avoid shell escaping issues:

cat > /tmp/seedance_body.json << 'ENDJSON'
{
  "prompt": "Design a motion-graphics style ad using glossy liquid glass design language. Pure black background. @Image1 is [describe]. @Image2 is [describe]. @Image3 is [describe]. The elements become translucent glass with reflections and refractions. Multiple camera angles. No text, no logos, no words. Style: liquid glass morphism, Apple Vision Pro aesthetic, premium 3D depth, self-luminous forms on absolute black.",
  "image_urls": [
    "https://your-upload-url/screenshot1.png",
    "https://your-upload-url/screenshot2.png",
    "https://your-upload-url/screenshot3.png"
  ],
  "resolution": "720p",
  "duration": "4",
  "aspect_ratio": "16:9",
  "generate_audio": false
}
ENDJSON

curl -s -X POST "https://fal.run/bytedance/seedance-2.0/reference-to-video" \
  -H "Authorization: Key $FAL_KEY" \
  -H "Content-Type: application/json" \
  -d @/tmp/seedance_body.json

Step 4: Download the Result

The response contains a video URL:

{
  "video": { "url": "https://v3b.fal.media/files/.../video.mp4" },
  "seed": 42
}

Download immediately — URLs may expire.

curl -sL -o "output_video.mp4" "THE_VIDEO_URL"

Step 5: Post-Production (Recommended)

Overlay the logo and any text in your video editor. The motion graphics are the visual base — branding goes on top in post.

Text-to-Video (No References)

For apps/projects where you don't have screenshots (GitHub repos, docs pages, abstract concepts), use text-to-video with the timestamped approach:

Ultra-sleek motion design sequence. Pure black background throughout.
Glossy liquid glass design language.

0-1s [FORM]: A translucent glass [object] materializes at center,
light refracting through it.

1-2s [CONNECT]: Glass connections form between elements, each a thin
glass tube filling with luminous light.

2-3s [REVEAL]: The structure rotates revealing depth and glass
reflections.

3-4s [GLOW]: A pulse of warm light passes through the entire glass
structure.

Style: liquid glass morphism, premium 3D depth, self-luminous forms
on absolute black.

Endpoint: https://fal.run/bytedance/seedance-2.0/text-to-video (Pro)

Same parameters minus image_urls, video_urls, audio_urls.

Image-to-Video (Single Image Animation)

Animate a single still image into cinematic video. Unlike reference-to-video (which blends multiple images into motion graphics), image-to-video takes one image and brings it to life with motion described in the prompt.

Endpoint: https://fal.run/bytedance/seedance-2.0/image-to-video

Parameters

Key Differences from Reference-to-Video

• Single image_url (string) instead of image_urls (array)
• end_image_url — optional end frame for controlled transitions (reference-to-video doesn't have this)
• Lip-sync support — audio generation includes lip-synced speech when faces are present
• No @Image1/@Image2 notation — just one image as the starting frame

When to Use Image-to-Video vs Reference-to-Video

Usage Example

cat > /tmp/seedance_i2v.json << 'ENDJSON'
{
  "prompt": "The dashboard interface comes alive — UI elements gently float and shift, glass reflections ripple across panels, accent lighting pulses through the layout. Smooth cinematic camera drift. No text, no logos.",
  "image_url": "https://your-upload-url/screenshot.png",
  "resolution": "720p",
  "duration": "4",
  "aspect_ratio": "16:9",
  "generate_audio": false
}
ENDJSON

curl -s -X POST "https://fal.run/bytedance/seedance-2.0/image-to-video" \
  -H "Authorization: Key $FAL_KEY" \
  -H "Content-Type: application/json" \
  -d @/tmp/seedance_i2v.json

Start + End Frame Example (A→B Transition)

Animate from one state to another — e.g. light mode to dark mode, empty dashboard to populated:

cat > /tmp/seedance_i2v_transition.json << 'ENDJSON'
{
  "prompt": "The interface smoothly transitions — colors shift, panels rearrange, elements morph fluidly from the starting layout to the final layout. Cinematic, smooth, continuous motion.",
  "image_url": "https://your-upload-url/state_before.png",
  "end_image_url": "https://your-upload-url/state_after.png",
  "resolution": "720p",
  "duration": "6",
  "aspect_ratio": "16:9",
  "generate_audio": false
}
ENDJSON

curl -s -X POST "https://fal.run/bytedance/seedance-2.0/image-to-video" \
  -H "Authorization: Key $FAL_KEY" \
  -H "Content-Type: application/json" \
  -d @/tmp/seedance_i2v_transition.json

Prompting Tips for Image-to-Video

• Describe the motion, not the image — the model already sees the image. Focus on what should move, how, and where.
• "Cinematic camera drift" works well for subtle product shots
• "Elements gently float and shift" for UI animation
• For lip-sync: describe the speech/emotion in the prompt, enable generate_audio: true
• Same negative guidance applies: "No text, no logos, no words" to prevent garbled text
• The end_image_url feature is powerful for before/after reveals, mode switches, or showing a workflow progression

Prompt Patterns Library

Floating Desktop Screens (Best for SaaS/Apps)

Design a motion-graphics style ad based on these images, using
multiple camera angles — close-ups that highlight the glass
reflections, various tap and click interaction animations, and a
focus on this glossy glass UI design language. @Image1 is [describe].
@Image2 is [describe]. @Image3 is [describe]. Show the desktop
screens floating in 3D space on a pure black background, tilted at
slight angles like a MacBook product shot. The UI elements on screen
become translucent glass with reflections and refractions. No text,
no logos, no words. Style: liquid glass morphism, Apple product ad,
premium desktop app showcase, self-luminous forms on absolute black,
[color] accent lighting.

Glass UI Ad (Based on LEPADPHONE Style)

Design a motion-graphics style ad based on this image, using multiple
camera angles — close-ups that highlight the glass reflections,
various tap/click interaction animations, and a focus on this glossy
"glass UI" design language. @Image1 is [describe the app/interface].
The interface elements become translucent glass panels with
refractions. No text, no logos. Style: liquid glass morphism, premium
depth, self-luminous forms on absolute black.

Floating Desktop Screens — Rubric (Tested)

Used for the Rubric dashboard promo. 3 refs (Flows + Skill Tree + Agent Icons), Pro 720p 4s, audio on.

Design a motion-graphics style ad based on these images, using
multiple camera angles — close-ups that highlight the glass
reflections, various tap and click interaction animations, and a
focus on this glossy glass UI design language. @Image1 is a desktop
application dashboard showing agent workflow connections on a
hexagonal grid. @Image2 is a desktop application showing a network
graph visualization with connected nodes. @Image3 is a desktop
application showing a grid of colorful pixel art icons. Show the
desktop screens floating in 3D space on a pure black background,
tilted at slight angles like a MacBook product shot. The UI elements
on screen become translucent glass with reflections and refractions.
No text, no logos, no words. Style: liquid glass morphism, Apple
product ad, premium desktop app showcase, self-luminous forms on
absolute black, orange accent lighting.

iPad Product Reveal — Rubric (Tested)

Real device with glass screen content. 2 refs (hero + Flows), Pro 720p 4s, audio on. Key detail: explicitly call out "real solid device" so the iPad frame stays solid while only the screen gets glass effect.

Design a motion-graphics style ad using glossy liquid glass design
language. Pure black background, nothing else. A realistic iPad Pro
with visible black bezels and physical frame floating in empty black
space, tilted at a cinematic angle like an Apple product shot. The
iPad is a real solid device with clear borders and edges, not made
of glass. Only the screen content has the glass effect. The iPad
screen displays @Image1 and @Image2 — a dark dashboard with orange
branding, hexagonal grid, and agent flow connections. The UI elements
on the screen are translucent glass with reflections and refractions.
Light catches the physical edges of the iPad frame. The device
slowly rotates. No text, no logos, no words. Style: Apple keynote
product reveal, real device with glass UI on screen, premium 3D
depth, floating in absolute black void, orange accent lighting.

Abstract Glass Network — Rubric (Tested)

No device frame — pure abstract glass. 3 refs (Flows + Icons + Skill Tree), Fast 720p 4s, no audio.

Design a motion-graphics style ad using glossy liquid glass design
language. Pure black background. @Image1 is a dark dashboard with
hexagonal grid and agent connections. @Image2 is a grid of colorful
pixel art bot icons. @Image3 is a network graph of agents with
radiating connections. The hexagonal grid from Image 1 becomes
translucent glass honeycomb, each cell refracting light. Colorful
bot icons from Image 2 appear as glass tiles catching light. The
network from Image 3 pulses with glass fiber-optic connections,
light flowing between nodes. Multiple camera angles: close-up on
glass reflections, pull back to reveal the full network. Style:
liquid glass morphism, Apple Vision Pro aesthetic, premium 3D depth,
self-luminous forms on absolute black, orange accent lighting.

Abstract Motion Identity (Based on @oggii_0 Style)

Ultra-sleek motion design sequence. Pure black background throughout.
White-on-black minimal aesthetic. No cuts — one single continuous
unbroken visual evolution. Locked-off static center frame, no camera
movement. All motion occurs within the subject. Self-luminous white
forms on absolute black, no grain, no shadows.

0-2s [ORIGIN]: [Starting visual]
2-4s [TRANSFORM]: [Transformation]
4-6s [REVEAL]: [Reveal]
6-8s [CLOSE]: [Closing visual]

Style: [brand] motion identity, mathematical precision.
Mood: calm, inevitable, intelligent, premium.

Batch Generation Workflow

When generating promos for multiple projects at once, follow this process:

Step 1: Assess Each Project

For each app/project, determine:

• Do screenshots exist? Website or App Store → reference-to-video. GitHub repo or docs-only → text-to-video.
• What's the visual anchor? Dark UI = strong liquid glass match. Colorful icons = glass tiles. Geometric layout = glass honeycomb.
• What aspect ratio? Phone app → 9:16. Desktop/SaaS → 16:9. Abstract → 16:9.

Step 2: Test Batch (Cheap)

Run all projects at Fast tier, 480p, 4 seconds first. This keeps each test under ~$0.81. For a 5-project batch, total test cost is ~$4-5.

Step 3: Review & Promote Winners

Watch all test outputs. The ones that work well, re-generate at Pro tier, 720p, longer duration (8-10s). This is where you spend the real budget.

Step 4: Post-Production

Overlay logos and text in your video editor. The Seedance output is the visual base — branding goes on top in post.

Example: 5-Project Test Batch

All Fast, 480p, 4s (except iOS app at 6s for more screen time). Total: ~$5.50.

Cost Estimation

Before generating, estimate your cost:

Quick Formula

Cost = duration_seconds × $0.30 (Pro) or $0.24 (Fast) at 720p

480p is roughly 30% cheaper than 720p.

Estimation Table

Before You Generate Checklist

1. Pick your tier — Fast for testing, Pro for production
2. Pick your resolution — 480p for tests, 720p for keepers
3. Pick your duration — 4s minimum, 10s sweet spot, 15s max
4. Count your refs — 1 image per 2 seconds of video (max 5)
5. Audio on or off? — Same cost either way, but abstract visuals may trigger content policy. Default off for tests.
6. Multiply it out — duration × rate = cost per video

Gotchas & Known Issues

1. Text rendering is bad — never ask Seedance to render text. Overlay in post.
2. Audio content policy — abstract visuals may trigger false positives. Default to generate_audio: false, enable when needed.
3. People's faces in screenshots — may trigger "likenesses of real people" content policy. Crop faces out of screenshots.
4. Shell escaping breaks JSON — always use temp JSON files with cat > /tmp/file.json, never inline JSON in curl commands.
5. Kie AI upload URLs expire after 3 days — re-upload if generating later.
6. Max resolution is 720p — no 1080p. Upscale in post if needed.
7. Pro vs Fast — Pro is only ~25% more expensive but noticeably better. Default to Pro.
8. 7+ reference images — quality degrades. Stick to 5 or fewer.

Auth Header

Fal AI uses Authorization: Key (not Bearer):

Authorization: Key YOUR_FAL_KEY

Community Prompts

For inspiration beyond liquid glass, check the community collection:

awesome-seedance-2-prompts — 1,719 curated Seedance 2.0 prompts covering cinematic, anime, commercial, and experimental styles. Includes video examples and a searchable web gallery at youmind.com.

Credits

Built by RoboNuggets. Tested extensively with Rubric dashboard screenshots, App Store scraping, and liquid glass motion graphics.

Seedance 2.0 is by ByteDance. Fal AI is the API provider.