muapi-seedance-2
Expert Cinema Director skill for Seedance 2.0 (ByteDance) — high-fidelity video generation across Chinese, Global, and VIP tiers. Supports text-to-video, image-to-video, first-last-frame, omni reference, character training, omni-reference training, video editing, and watermark removal.
npx skills add SamurAIGPT/Generative-Media-Skills --skill seedance-2 --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
What it does
Guides an AI agent to perform Seedance 2.0 cinema tasks across multiple tiers, including text-to-video, image-to-video, frame interpolation, omni reference, character training, video editing, and watermark removal. Emphasizes technical cinematography, physics directives, and precise camera grammar to generate high-fidelity videos.
How it works
The skill provides a structured set of capabilities: text-to-video, image-to-video, extend video, first/last frame interpolation, omni reference, omni-reference training, character sheet creation, video editing, and watermark removal. It defines tier-specific constraints (Chinese, Global, VIP), input limits (images, videos, audio), and output expectations (4–15 seconds, 480p–720p with auto audio). It uses a six-component prompt framework (Scene, Subject, Action, Camera, Audio, Pacing/Style) to build Director Briefs, enforces a single-beat-per-segment rule for timing, and requires an Audio directive in prompts. It also includes a comprehensive reference system with tags for images, videos, audio, and character references, plus example templates and capability-specific patterns for consistency, camera movements, extension, editing, and effect aims. Commands and file references are presented as examples for integration with generation scripts (e.g., bash scripts/generate-seedance.sh) and explicit reference tokens (e.g., @image1, @Video1).
When to use it
Use when you need cinematic video generation or editing with Seedance 2.0 across supported tiers. Trigger prompts that specify one of the listed core competencies (e.g., text-to-video, image-to-video, video editing, watermark removal) and adhere to tier constraints and required Audio directive. Apply the six-component Director Brief structure and time-segment prompts for 10s+ videos.
What it can touch
The skill references tools for execution, including the @ Reference System for assigning assets and the Script-based generation workflows (e.g., bash scripts/generate-seedance.sh). It supports inputs of images (up to 9 per i2v/omni, formats jpeg/png/webp), audio (mp3/wav), and video (mp4/mov, with capacity limits). It requires explicit asset tagging and character references when provided. It notes that Seedance 2.0 can generate audio natively and requires an Audio directive in every prompt.
Caveats
Restrictions include no realistic human faces in uploads (except in specific training modes), and certain modes require prior seeds or tier specifications (e.g., --mode extend requires a request_id from prior jobs; first-last requires higher tiers). Global/VIP omni does not support video references (only images and audio). Output durations and quality parameters vary by tier and are constrained to 4–15 seconds with 480p–720p output.
# 🎬 Seedance 2.0 Cinema Expert **The definitive skill for "Director-Level" AI video orchestration.** Seedance 2.0 is not a descriptive model; it is an *instructional* model. It responds best to technical cinematography, physics directives, and precise camera grammar. ## Core Competencies 1. **Text-to-Video (t2v)**: Generate cinematic video from a Director Brief — Chinese, Global, or VIP tier. 2. **Image-to-Video (i2v)**: Animate 1–9 reference images — Chinese, Global (smart mode), or VIP tier. 3. **Video Extension (extend)**: Seamlessly continue an existing Seedance 2.0 video (Chinese tier). 4. **First & Last Frame (first-last)**: Interpolate a fluid video between a start image and end image (Global/VIP). 5. **Omni Reference (omni)**: Full multimodal reference with images + audio + character refs (all tiers). 6. **Omni Reference Training (omni-train)**: Train a custom persistent character for identity-consistent generation. 7. **Character Sheet (character)**: Build a reusable character from 1–3 images (Chinese tier). 8. **Video Edit (video-edit)**: Edit an existing video with a prompt + optional reference images (Chinese tier). 9. **Watermark Removal (watermark-remove)**: Strip Se
- Core Competencies
- 🏷️ Tiers
- 📥 Input Limits
- ⚠️ Restrictions
- 🔗 Core Syntax: The @ Reference System
- Chinese Tier (i2v, omni)
- Global/VIP Omni (omni-reference-no-video / vip-omni-reference)
- Character References (all tiers)
- Role Assignment Table
- Multi-Reference Combination
- 🏗️ Technical Specification: The Director Brief
- Time-Segmented Prompts (Recommended for 10s+ videos)
- Negative Prompting
- 🎥 Camera Language Reference
bash scripts/generate-seedance.sh \ trees rushing past in peripheral blur, rocky terrain below. Slight natural stabilization with wide-angle lens. Continuous forward motion, no cuts throughout. Trail opens into a clearing — mountain peak visible ahead. Smooth gimbal-stabilized descent along a sweeping lateral arc, dropping toward a rooftop terrace. Long shadows cast across building tops, warm light on ocean surface. High-speed dive closes in to product on the terrace — final frame settles into a medium close-up. Chinese tier (default) — epic reveal Global tier — 21:9 cinematic, 12s VIP fast — square social format Chinese tier — animate with video/audio refs
What does the muapi-seedance-2 skill do?
Expert Cinema Director skill for Seedance 2.0 (ByteDance) — high-fidelity video generation across Chinese, Global, and VIP tiers. Supports text-to-video, image-to-video, first-last-frame, omni reference, character training, omni-reference training, video editing, and watermark removal.
How do I install it?
Run `npx skills add SamurAIGPT/Generative-Media-Skills --skill seedance-2 --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From SamurAIGPT/Generative-Media-Skills, a repository with 3,976 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.