Full-stack setup overview

The full-stack setup gives you the whole platform on your machine: the Admin UI, MongoDB, RabbitMQ, the Orchesty backend services, and one or more workers. Nothing leaves your laptop.

When to choose this #

• You need offline development.
• You want the Admin UI, queues, and persistence on your own machine without depending on an existing instance.
• You want to control every component, including the Admin UI version, secret encryption, and the message broker.
• Your organisation requires on-prem data flow even at development time.
• You are evaluating Orchesty end-to-end and want to see all moving parts.

What you get #

The project skeleton (Orchesty/orchesty-skeleton) is the meta-repository that wires the platform together. It contains:

orchesty-skeleton/
  docker-compose.yml       # the platform: Admin UI, backend, Mongo, RabbitMQ, supporting services, worker
  Makefile                 # init-dev, docker-up-force, docker-stop, docker-down-clean
  AGENTS.md                # setup workflow for AI editors (Cursor, Claude Code, ...)
  worker/                  # your Node.js worker source, with its own Makefile
  topology/                # exported topology definitions, auto-installed by init-dev
  .env.dist                # template environment variables (UID/GID, DEV_IP, secrets)

The worker/ subfolder is your project. It already contains a wired SDK install, a Dockerfile, and a Makefile that proxies common commands into the running container. Everything outside worker/ is the platform itself, kept stable across updates.

Prerequisites #

• Docker and Docker Compose (Docker Desktop on macOS / Windows, native on Linux).
• make.
• 4 GB of free RAM for the stack.

The worker container brings its own Node.js runtime, so you do not need Node.js installed on the host unless you want to run lints/tests outside Docker.

What's next #

1. Install the full stack.
2. Your first worker.
3. Your first custom node.