Integration & Runtime

This manual is for engineers driving Composer from outside — backend services, scripts, control surfaces, custom front-ends. It complements the per-component reference manuals: the Inputs, Operators, and Targets describe what's available; this manual shows how to reach in and control any of it programmatically.

For the operational side of running the headless playback engine — install, command-line arguments, configuration, containers, logging — see the dedicated Composer Runtime manual.

Composer offers several complementary integration paths — pick the right tool for the job:

• HTTP API — direct HTTP requests against a running project for single actions, queries, and orchestration from any backend.
• Runtime API — separate HTTP endpoints for the runtime process itself — start, stop, reload projects, manage the runtime's lifecycle from outside.
• Connectors — GUI-configured action sequences triggered by name or value over the HTTP API. No coding required for the Composer side.
• Batches — control groups of layers tagged with a shared name. Documented as part of Scenes and layers in the User Guide.
• Script Engine — JavaScript runtime embedded per project for conditional logic and event handlers.
• WebSockets — real-time bidirectional event stream for live updates and continuous data.

These tools combine: Connectors can trigger scripts and Batches; the HTTP API can invoke Connectors; the WebSocket stream surfaces state changes that scripts react to; the Runtime API sits one level above and controls which project the HTTP API is talking to. There's often more than one way to achieve the same result — the "best" approach depends on your team's preferences, where the source of truth lives, and how much logic you want inside Composer vs. outside.