A sophisticated AI video generation skill that transforms text, images, and multimodal references into cinematic 4-15 second clips using the Seedance 2.0 API.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install seedance2-video-gen
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install seedance2-video-gen using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Seedance 2.0 Video Generation?
The Seedance 2.0 video generation skill is a powerful extension for AI agents within the Openclaw Skills ecosystem. It enables developers and creators to programmatically generate high-fidelity video content by interfacing with the EvoLink.ai API. Whether you are starting from a simple text prompt, animating a static image, or using a complex reference video for motion transfer, this skill handles the heavy lifting of parameter optimization and API communication.
Built for precision, it supports specialized syntax for granular control over characters, camera movements, and synchronized audio. By integrating this skill, users can automate the creation of marketing assets, social media content, or cinematic sequences with consistent quality and professional-grade resolution up to 720p.
Seedance 2.0 Video Generation Use Cases
- Creating 5-15 second cinematic video clips from natural language descriptions.
- Animating static product images or character portraits into dynamic scenes.
- Transferring camera movements and character actions from a reference video to a new AI-generated subject.
- Synchronizing AI-generated visuals with background music and sound effects automatically.
- Rapid prototyping of video concepts using the high-speed fast-track models.
How Seedance 2.0 Video Generation Works
- Requirement Analysis: The agent parses user input to identify the core subject, desired duration (4-15s), and aspect ratio (e.g., 16:9, 9:16).
- Model Selection: Based on the input type (text-only, image-heavy, or multimodal), the skill selects the optimal model from the seedance-2.0 suite.
- Prompt Engineering: The skill constructs a structured prompt using specialized @material syntax to link images, videos, and audio files as context.
- API Parameter Generation: A valid JSON payload is generated according to the EvoLink API specifications, including quality settings and audio toggles.
- Execution & Retrieval: The skill provides a complete execution command (curl/Python) to trigger the generation task and retrieve the final video link.
Seedance 2.0 Video Generation Setup
To use this skill within Openclaw Skills, you must first obtain an API key from EvoLink.ai. Set your environment variable as follows:
export SEEDANCE_API_KEY="your_api_key_here"
Ensure your local environment can reach the EvoLink API endpoints at [https://api.evolink.ai/v1/](https://api.evolink.ai/v1/). No additional local dependencies are required if using standard HTTP clients like curl.
Seedance 2.0 Video Generation Data Schema & Taxonomy
The skill organizes input and output data based on the following technical specifications:
Input Specifications
| Type | Format | Max Count | Size Limit |
|---|---|---|---|
| Images | jpeg, png, webp, bmp | 9 files | < 30MB |
| Video | mp4, mov | 3 files | < 50MB |
| Audio | mp3, wav | 3 files | < 15MB |
Output Taxonomy
- Resolution: 480p or 720p.
- Duration: 4-15 seconds (or adaptive).
- Aspect Ratios: 16:9, 9:16, 1:1, 4:3, 3:4, 21:9, and adaptive.
- Link Validity: Generated video URLs remain active for 24 hours.
Seedance 2.0 Video Generation Advanced Features
- @Material Syntax: Allows precise control by referencing specific assets (e.g., '@image1 as the first frame').
- Timeline Segmentation: Supports complex scenes by breaking prompts into time blocks (e.g., '0-3s: zoom in, 3-6s: character waves').
- Motion Reference: Ability to extract camera paths from reference videos to ensure stable and realistic movement.
- One-Shot Mode: Specialized prompting for continuous, no-cut cinematic shots.
- Video Extension: Capabilities to extend existing videos by describing the state of the final frame.
SKILL.md
Loading
Related Openclaw Skills
Desktop Automation Pro
A comprehensive cross-platform toolkit for automating web browsers, mobile devices, and native desktop applications via GUI interaction.
Automation Workflows
Design and implement high-impact automation workflows to eliminate repetitive manual tasks and scale your business operations.
skill-dep-fixer
A specialized Node.js CLI tool designed to scan OpenClaw skill files and automatically resolve missing system, npm, pip, and Homebrew dependencies.
Seedance Prompt Wizard
A guided conversational assistant that transforms creative ideas into technical, production-ready prompts for Seedance 2.0 video generation.
Codex Swarm
A high-performance orchestration framework for spawning parallel OpenAI Codex agents to handle complex, multi-task coding projects simultaneously.
302.AI API Integration Assistant
A specialized developer tool for discovering and integrating over 1400 AI services from 302.AI directly into any codebase.
Featured*
