seedance-video-creator_skill

This skill helps you generate professional storyboard prompts and directly produce seedance videos with multi-modal inputs and automated downloads.
  • Python

122

GitHub Stars

2

Bundled Files

3 weeks ago

Catalog Refreshed

2 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstart where the catalogue uses aiagentskills.

npx veilstart add skill wwwzhouhui/skills_collection --skill seedance-video-creator

  • README.md6.1 KB
  • SKILL.md17.1 KB

Overview

This skill converts creative ideas into professional storyboard prompts and can directly generate videos using the Seedance 2.0 family via a jimeng-compatible API. It supports multimodal references (images, short videos, audio), guided or quick workflows, and automatic download of generated video files. Use it to get concise Seedance-ready prompts or to produce finished short videos with minimal manual steps.

How this skill works

I ask focused questions to capture story, duration, style, camera actions, and reference files, then produce a time-coded storyboard prompt formatted for Seedance 2.0. If the user opts in, I call a jimeng-free-api-all endpoint with the prompt and uploaded files, poll until completion, extract the returned video URL, and optionally download the video to a local path. The tool handles text-only, local file multipart uploads, and URL-based file references.

When to use it

  • You want a professional, time-coded storyboard prompt tailored to Seedance 2.0 syntax.
  • You have reference images/videos and want them integrated into a short generated video.
  • You prefer step-by-step guided creation when you are unsure about visual details.
  • You already have a clear prompt and want fast generation with minimal interaction.
  • You need automatic download and a ready-to-share MP4 file after generation.

Best practices

  • Provide a one-sentence story summary and preferred duration (4/5/10s) up front.
  • Upload prioritized references (max 12 files total; images ≤9, videos ≤3, audio ≤3).
  • Use @1/@2 placeholders in prompts to map to uploaded files in order.
  • Keep prompts under ~500 Chinese characters (or equivalent) to avoid degradation.
  • Set curl timeout to 300s when calling the API and download generated URLs promptly.

Example use cases

  • Create a 10s cinematic promo from concept to final MP4 using two reference photos.
  • Generate a quick 5s vertical social clip from a short text storyboard (text-only).
  • Produce a dance scene using a reference video for camera movement replication.
  • Build a product teaser with closeups and brand splash using product images.
  • Extend or edit an existing video by referencing @video1 and specifying modifications.

FAQ

Run the jimeng-free-api-all service (default port 8000) and obtain a sessionid from jimeng.jianying.com cookies or provide it as JIMENG_SESSION_ID.

Why did generation fail or return 401/429?

401 means session expired—get a fresh sessionid. 429 indicates quota limitations; wait, reduce usage, or top up credits.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational
seedance-video-creator skill by wwwzhouhui/skills_collection | VeilStrat