# Generative Media Skills

A library of more than 150 agent skills that teach AI coding assistants how to plan, generate, and check image, video, audio, and voice production work.

Generative Media Skills is a collection of agent skills, organized into 25 categories, for producing and validating image, video, audio, voice, music, 3D, avatar, and other generative media inside AI coding assistants such as Claude Code, Codex, GitHub Copilot, and Cursor. Rather than only generating content, the skills combine researched production knowledge, comparisons of current providers, models, gateways, and costs, with executable guardrails and technical checks.

A typical workflow moves from choosing the right system for a task, through directing the work with cinematography, sound, and editing craft, to building the actual deliverable, running deterministic checks on the output, such as inspecting media, captions, 3D assets, or loudness, and finally handling provenance, rights, and delivery requirements. Each skill's own documentation file explains when and how to use any bundled executable resources. It can also connect to a separate project for orchestrated media workflows.

Website: https://github.com/calesthio/generative-media-skills