Analysis updated 2026-05-18
Turn a rough action or story idea into a matched storyboard and video prompt pair.
Convert an emotional or romantic scene concept into a shot-by-shot cinematic previs breakdown.
Install reusable video prompting skills into Codex or Claude Code with one command.
| agi-ruby/ai-gpt_image2-seedance_2.0-video-skills | jeonghopark/collective-trajectories | patrickarlt/custom-element-decorators | |
|---|---|---|---|
| Stars | 10 | 10 | 10 |
| Language | JavaScript | JavaScript | JavaScript |
| Last pushed | — | — | 2015-10-24 |
| Maintenance | — | — | Dormant |
| Setup difficulty | easy | easy | moderate |
| Complexity | 1/5 | 4/5 | 2/5 |
| Audience | vibe coder | designer | developer |
Figures from each repo's GitHub metadata at analysis time.
This repository packages two reusable skills for AI assisted video creation, meant to be installed into AI coding tools like Codex or Claude Code. The skills do not generate video themselves. Instead they provide structured prompt templates and workflows that help you write better prompts for AI video generation tools, specifically GPT Image 2 and Seedance 2.0. The project exists because video prompts tend to fail in one of two ways: they are too vague, so the model invents its own pacing and shot order, or they are specific but one off and hard to reuse. These skills turn the process into a repeatable system instead. The first skill, storyboard-to-seedance-video, walks through a two stage pipeline. You start with an idea, and it helps generate a storyboard prompt for GPT Image 2, then turns that storyboard into a structured video prompt for Seedance 2.0. It includes prompt templates, a filling guide, output format references, and a worked kung fu action example. The second skill, romantic-ink-cinema-previs, targets emotional and stylized video concepts rather than literal plot description. It helps translate a feeling or relationship dynamic into specific cinematic terms such as camera choreography, body rhythm, spatial tension, and pacing. It suits romantic or poetic image to video work with soft light and energy effects. Both skills install through a command line tool using npx, targeting whichever AI environment you use, including a single skill install and an uninstall option. The project is written in JavaScript, does not call any video generation APIs itself, and focuses purely on prompt structure and reusable creative workflows.
Two installable prompt skills that turn a rough video idea into structured, reusable prompts for GPT Image 2 storyboards and Seedance 2.0 video generation.
Mainly JavaScript. The stack also includes JavaScript, Node.js, npx.
No license information is stated in the README.
Setup difficulty is rated easy, with roughly 5min to a first successful run.
Mainly vibe coder.
This repo across BitVibe Labs
Verify against the repo before relying on details.