How awesome-genmedia/skills works
This repository is a catalog of generative-media skills for AI agents, rather than an application runtime. Users invoke a skill for a media task; the README documents use-case skills with each::sense’s model-selecting API and direct model access through the Prediction API. The catalog also includes domain-specific categories and individual model skills. No source excerpts were provided, so relationships beyond these documented entry points are omitted.
Main entry points: image-generation/SKILL.md Image Skills · image-editing/SKILL.md Image Editing · video-generation/SKILL.md Video Skills · song-generation/SKILL.md Song Generation
Components 21
Core Skills
- Image Skills
image-generation/SKILL.md - Image Editing
image-editing/SKILL.md - Image Upscaling
image-upscaling/SKILL.md - Background Removal
background-removal/SKILL.md - Face Swap
face-swap/SKILL.md - Video Skills
video-generation/SKILL.md - Video Editing
video-editing/SKILL.md - Song Generation
song-generation/SKILL.md - Music Generation
music-generation/SKILL.md - Lyrics Generation
lyrics-generation/SKILL.md - Voice Generation
voice-generation/SKILL.md - Text to Speech
text-to-speech/SKILL.md - Speech to Text
speech-to-text/SKILL.md - Sound Effects
sound-effects/SKILL.md
Skill Catalog
- Domain Skills
categories - Model Skills
models
API Access
- each::sense API
- Prediction API
Other components
- Agent User
- each::labs
- Generated Media
Connections 25
- Agent User → Image Skills: invokes
README.md - Agent User → Image Editing: invokes
README.md - Agent User → Image Upscaling: invokes
README.md - Agent User → Background Removal: invokes
README.md - Agent User → Face Swap: invokes
README.md - Agent User → Video Skills: invokes
README.md - Agent User → Video Editing: invokes
README.md - Agent User → Song Generation: invokes
README.md - Agent User → Music Generation: invokes
README.md - Agent User → Lyrics Generation: invokes
README.md - Agent User → Voice Generation: invokes
README.md - Agent User → Text to Speech: invokes
README.md - Agent User → Speech to Text: invokes
README.md - Agent User → Sound Effects: invokes
README.md - Agent User → Domain Skills: selects
README.md - Agent User → Model Skills: selects
README.md - Image Skills → each::sense API: uses for requests
README.md - Image Editing → each::sense API: uses for requests
README.md - Video Skills → each::sense API: uses for requests
README.md - Song Generation → each::sense API: uses for requests
README.md - each::sense API → each::labs: sends requests
README.md - Model Skills → Prediction API: direct model access
README.md - Agent User → Prediction API: submits model request
README.md - Prediction API → each::labs: sends prediction
README.md - each::labs → Generated Media: generates
README.md