Embedding Dusk

This is Dusk's whole reason for existing: you embed it into software you already ship, and that software's devices become a managed fleet. No separate service to stand up, no rewrite - you link Dusk in, call one function, and you're done.

Picture the software inside a media player, a fleet of robot vacuums, or a rack of build machines. Each is a device someone needs to watch and operate. Embed Dusk and every one of them becomes a node you can get analytics from, diagnose, and drive - at any scale, on anything from a microcontroller to a supercomputer.

How you embed it

The universal integration is a C library. Dusk's node compiles to a static library with a single-call entry point, so it drops into an app written in practically any language. See The C library.

TODO - embedding is more than the C library. The C lib is the first integration, not the only one. Native bindings for major languages and frameworks are on the roadmap, so embedding a node becomes idiomatic in each ecosystem rather than an FFI call. These pages will grow as those land.

TODO - the connection layer. How an embedded node is then discovered and securely reached by a fleet (identity, encryption, transport) is still being built. Linking the library and starting a node works today; the secure fleet-wide connection story is in progress.

Why this is the headline

Most users never write a Dusk program or port it to new hardware - that's the framework, the advanced path. The common case is far simpler: take what Dusk already does, and put it inside what you already have.