seedance-2-video-gen

What This Skill Does

The seedance-2-video-gen skill enables OpenClaw, Claude Code, and Cursor users to leverage the powerful ByteDance Seedance 2.0 model for high-quality video synthesis. By integrating directly with the EvoLink API, this skill abstracts away the complex infrastructure required to run diffusion-based video models. It supports both text-to-video transformations and image-to-video animations, automatically generating synchronized audio tracks including voice-overs, sound effects, and background music. The skill is designed with a conversational interface that guides users through the configuration process, ensuring that parameters such as duration, resolution, and artistic intent are gathered efficiently without overwhelming the user.

Installation

To install the skill, use the following command in your terminal within your OpenClaw environment:

clawhub install openclaw/skills/skills/evolinkai/seedance-2-video-gen

After installation, the agent will verify your EVOLINK_API_KEY. If it is missing, follow the provided link to sign up at evolink.ai. Once the key is configured, the skill remains active in your session, waiting for your creative prompts.

Use Cases

• Prototyping Content: Rapidly visualize video concepts for marketing or storytelling without expensive production cycles.
• Asset Generation: Create background stock footage, short cinematic clips, or social media content directly from natural language.
• Iterative Animation: Use reference images to animate static assets for game development or design mockups.
• Multimedia Creation: Generate fully rendered clips with embedded, context-aware audio, reducing the need for post-production video editing.

Example Prompts

1. "I want to generate a 10-second cinematic video of a cyberpunk city street in the rain with synth-wave background music at 1080p resolution."
2. "Can you create a 6-second animation based on this image of a cat? I'd like it to look like it's jumping through a portal."
3. "Help me make a 4-second video clip of a sunrise over a desert, I want to include ambient wind sounds and soft background music."

Tips & Limitations

• Duration: The model supports clips between 4 and 12 seconds. Requests outside this range will be normalized or rejected.
• Resolution: You can choose between 480p, 720p, and 1080p. Higher resolutions require more processing time.
• Audio: Audio generation is a core capability; specify if you want specific elements like 'voice-over' vs 'ambient SFX' for better results.
• Reference Images: For image-to-video, ensure your assets are under 30MB and in standard formats (PNG/JPEG). You may provide up to 9 reference frames to guide the visual consistency of the generation.