What problem does it solve? Creating AI-generated content like YouTube videos, product demos, or talking-head clips requires chaining many separate models (image generation, animation, TTS, merging) by hand, which is slow and error-prone. This Skill provides ready-made pipeline patterns that orchestrate these steps through the inference.sh belt CLI. ## Core Features & Use Cases - Pipeline Patterns: Predefined chains such as Image -> Video -> Audio, Script -> Speech -> Avatar, and Research -> Content -> Distribution. - Complete Workflows: Step-by-step bash commands for YouTube Shorts, talking-head videos, product demos, and blog-to-video conversion using FLUX, Wan 2.5, Kokoro TTS, OmniHuman, and media-merger apps. - Building Blocks Reference: Tables mapping each pipeline stage (script, visual, animation, audio, post-production) to the appropriate inference.sh app. - Use Case: A marketer needs a product showcase video; the Skill walks through generating a product image with FLUX, animating it with Wan 2.5, upscaling with Topaz, and merging background music. ## Quick Start Ask the AI to create a YouTube Short about a topic by generating a script, voiceover, background image, and animated video, then merging them with the media-merger app.
