Author with the Composition API
The Composition API is the product-level authoring surface. It produces the same Project v1 used by media import, Session, persistence, preview, and export.
Create a Composition
Section titled “Create a Composition”Create a composition with a stable ID, canvas size, rational frame rate, sample rate, and duration. Add ordered visual, audio, and caption layers. Use the time helpers for every placement and duration.
Add content
Section titled “Add content”Clips can describe imported media, images, styled text, vector shapes, captions, solid/generated content, and nested sequences. Visual properties include transform, opacity, crop/mask, blend behavior, animation, and enabled effects. Audio clips carry source mapping, gain, fades, and mixer relationships.
Reuse Materials and transitions
Section titled “Reuse Materials and transitions”Reference installed Materials by exact package/definition identity and provide validated parameter values and resource bindings. An effect belongs to its host item; a transition binds two compatible visual inputs over a valid interval.
Build and load
Section titled “Build and load”build() creates a plain Project and validates the authored structure. Bind its asset IDs to a
media provider, create a Session, and call loadProject(). The Session performs the full untrusted
input, reference, time, ownership, Material, and execution validation.
advanced()
Section titled “advanced()”Use the advanced escape hatch only for Project features not yet represented by the high-level builder. Keep IDs and ownership stable, return canonical JSON, and treat validation failure as a construction error. Do not use it to bypass semantic commands during interactive editing.
Continue with Importing media and Timeline editing.