AI Agent Hub
Back to skills
Seedance 2.0 Open Source icon

Seedance 2.0 Open Source

Design & Media Updated 2026.08.30

Paste the following prompt into your AI chat to install this skill:

Please install @user_e942ebfc/seedance2-new according to https://skillhub.cn/install/skillhub.md.

About this skill

The Challenge of Rapid Video Generation from Text

When creating content for social media, product demos, or educational materials, there is often a need to quickly convert text descriptions into video assets. Traditional approaches may involve professional video editing tools, complex API integration (such as applying for Volcengine whitelists, configuring AK/SK keys), or subscription-based services. These processes are time-consuming and have high barriers to entry, making it difficult for rapid iteration and testing of creative ideas.

How Seedance 2.0 Open Source Works

This skill wraps the Volcengine Seedance 2.0 model and simplifies API authentication through the redfox.hk platform, allowing developers to generate videos with a single command.

  • Core Capabilities:
  • Text-to-Video: Input Chinese or English prompts to automatically generate MP4 videos with synchronized audio.
  • Parameter Control: Adjust resolution (--resolution 480p/720p/1080p), aspect ratio (--ratio 16:9/9:16), and duration (--duration 4-15 seconds) via command-line arguments.
  • Virtual Avatar Reference: Use asset:// URLs to reference preset virtual avatars without uploading real human images.
  • Task Management: Submit a task to get a taskId, then query progress and results.

  • Key Workflow:
    1. Set the REDFOX_API_KEY environment variable (obtained by registering on redfox.hk).
    2. Run a command like python3 videogen.py "A cat running on grass".
    3. The script automatically polls the task status and downloads the video to the specified directory upon completion.

Boundaries and Considerations

  • Generation Time: Video generation typically takes 3–15 minutes, possibly longer for complex scenes; the script waits automatically, but you can use --no-download to only submit without waiting.
  • Content Restrictions: Does not support uploading reference images with real human faces; to specify a person, describe in the prompt or use the virtual avatar library.
  • API Dependency: Requires a valid REDFOX_API_KEY with sufficient balance; if generation fails, you can re-query with --task-id.
  • Output Format: Saves as MP4 by default; customize the directory with -o.

Use Cases

  • During TikTok or Xiaohongshu operations, need to quickly convert a text description into a 9:16 vertical video asset for daily content posting.
  • Product managers preparing for product roadshows use the tool to generate concept videos from feature descriptions, embedding them in presentations to visually demonstrate design ideas.
  • Marketing personnel batch-generate brand promotional shorts by fixing prompts and parameters (e.g., 16:9 ratio, 1080p resolution) to ensure consistent visual style.
  • Designers in the creative validation phase describe abstract concepts in natural language to generate video previews, assessing feasibility and iterating directions.

Best For

  • Short video bloggers: need to generate multiple vertical video assets weekly to maintain social media account update frequency.
  • Startup product managers: need to create product concept videos within tight deadlines for internal reviews or investor demonstrations.
  • Brand content operators: manage corporate social media accounts and need to batch-produce promotional videos that align with brand tone.
  • Online education instructors: need to convert abstract knowledge points into visual clips to help students understand complex concepts.