generative-media

Paths below are relative to this skill's directory.

Scope

Calling a model that emits or reads pixels, video frames or audio samples: image generation and editing, video generation, speech synthesis, speech recognition, music, and local diffusion pipelines. Everything around that call — how the request is shaped, how the job is run, what the output costs, what must be stored, and what must be disclosed.

The same input must produce the same bytes. That is the line: work whose output varies with a model, a seed or a server-side version belongs here; work that is a pure function of its input belongs to the media-processing skill.

Not covered — do not answer from this skill: