media-generation

Create AI-driven videos by describing scenes in declarative JSX with Render and Clip workflows.

20|2|Updated Jan 22, 2026

Search Tags:#workflow #rendering #video #declarative #jsx #ai-video

Ready to run

npx skills add https://github.com/vargHQ/skills --skill media-generation-varghq

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: media-generation
Source: https://github.com/vargHQ/skills/tree/main/media-generation
Command: npx skills add https://github.com/vargHQ/skills --skill media-generation-varghq

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Create and orchestrate AI-driven videos by describing scenes in JSX, removing the need to hand-code complex workflows and synchronizing assets, timing, and effects automatically.

Core Features & Use Cases

Declarative scene composition using Render, Clip, Image, Video, Music, Speech, Title, and Grid components.
Automatic parallel rendering and caching to speed up iteration.
Provider-agnostic integration with fal, elevenlabs, and other backends for images, video, and audio.
Use cases include TikTok/YouTube Shorts promos, talking-head videos, product demos, and cinematic social content.

Quick Start

Describe your video as a JSX scene using Render, Clip, Image, Video, and Music components, then render it to a 1080x1920 TikTok-ready video.

Frequently Asked Questions about media-generation

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I generate AI videos using declarative JSX instead of imperative API calls?▼

AI video generation with declarative JSX involves describing scenes using components like Render, Clip, and Image, which automatically synchronizes assets and timing. This removes the need to hand-code complex workflows, handling parallel rendering and caching internally.

Can I use fal and elevenlabs backends together for rendering video and audio?▼

Yes, provider-agnostic integration supports combining fal and elevenlabs backends within the same JSX workflow. This allows you to seamlessly compose images, video, and speech across different providers without writing separate API synchronization logic.

What is the best way to create TikTok-ready vertical videos from text and images?▼

The best way to create TikTok-ready vertical videos is by describing your scenes declaratively in JSX. The workflow outputs a 1080x1920 format video automatically, utilizing built-in parallel rendering and caching to speed up the iteration process.

How does declarative scene composition handle parallel rendering and caching?▼

Declarative scene composition handles parallel rendering and caching automatically by orchestrating JSX components like Render, Clip, and Music. This built-in mechanism speeds up iteration by managing asset synchronization and rendering tasks without requiring manual API calls.

Do I need to import external libraries to compose AI video clips in my code?▼

No, you do not need to import external libraries to compose AI video clips. The workflow operates with zero required imports in your code, allowing you to directly declare scenes using Render, Clip, Image, Video, and Music components.

What types of social media content can I produce with a declarative video workflow?▼

A declarative video workflow produces social media content like TikTok and YouTube Shorts promos, talking-head videos, product demos, and cinematic social content. It achieves this by composing images, video, and audio through JSX scene descriptions.