zhanghaonan777

Seedance2 视频创意技能包:100+ 镜头词库、Seedance 2.0 全模态 API CLI,兼容 OpenClaw / Cursor / 任意 Agent 平台。AI Agent skill for ByteDance Seedance video generation. Creativity gate (memorability / surprise / emotion / narrative), zero-copy ideation from images, 100+ cinematography terms, Seedance 2.0 multimodal API CLI. Works with OpenClaw, or any agent platform.

23
3
100% credibility
Found Feb 27, 2026 at 17 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

A skill pack and simple tool that empowers AI assistants to generate creative videos from text, images, videos, or audio using ByteDance Seedance models.

How It Works

1
🔍 Discover the video magic

You stumble upon this fun skill that lets your AI buddy become a creative video director using special video-making powers.

2
📋 Get your video service ready

You sign up for the video creation service and connect it so everything can talk to each other smoothly.

3
📁 Add the skill to your AI helper

You place the skill folder into your AI friend's special tools area, like OpenClaw or Cursor.

4
💡 Dream up your video

You share a simple idea like 'a man running through a market' or add a photo, music, or clip, and the AI brainstorms exciting twists.

5
🎬 Let the AI create

The AI reviews for memorable moments, surprises, emotions, and story flow, then generates your video just the way you imagined.

🎉 Your video is ready!

You download the stunning video to your desktop, amazed at how the AI turned your idea into a professional clip.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 17 to 23 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is Seedance2-skill?

This Python project packs a CLI and agent skill for ByteDance Seedance video generation, letting AI agents handle multimodal inputs like text, images, video, and audio via the Seedance 2.0 API. It solves the problem of bland AI videos by enforcing a creativity gate—checking memorability, surprise, emotion, and narrative—while offering zero-copy ideation from images and a 100+ cinematography term library. Users get one-command generation with options for ratios, durations, motion replication, and beat-sync audio.

Why is it gaining traction?

It stands out with an autonomous creativity system that iterates prompts until they pass the gate, producing engaging outputs without fixed templates, unlike basic prompt wrappers. The CLI covers all Seedance models with no dependencies, supports webhooks and chaining via last-frame returns, and plugs into agent platforms like Cursor or OpenClaw. Developers dig the full multimodal API access and reference vocab for pro-level results.

Who should use this?

AI agent builders integrating video tools into Cursor or OpenClaw workflows. Content creators prototyping ByteDance clips from images or audio beats. Devs experimenting with agent-driven ideation for ads, social reels, or narrative shorts needing emotion and surprise.

Verdict

Grab it if you're in the ByteDance ecosystem—solid docs, MIT license, and zero deps make it dead simple to test, despite 17 stars and 1.0% credibility signaling early maturity. Skip for production until more adoption confirms reliability.

(178 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.