Skip to main content
Live. This area is documented as current, user-reliable behavior.

​
Goal

Give new users enough vocabulary to understand the rest of the docs.

​
Prerequisites

  • Read “What is StackShift” first

​
Workflow

1
Learn the difference between project deployments and stack deployments.
2
Treat templates as stack blueprints, not as a separate runtime engine.
3
Treat nodes as schedulable runtime hosts, not as dashboard-only records.
4
Treat operations as the cross-resource health and recovery surface.

​
Objects you manage

  • Project: an app deployment unit, usually GitHub- or image-based.
  • Stack: a Compose-defined service unit, often stateful or service-oriented.
  • Template: a versioned blueprint that renders into a stack.
  • Node: a connected machine running the StackShift agent.
  • Database: a managed database resource with its own lifecycle and recovery behavior.

​
Operational language

  • Placement: where a workload is scheduled.
  • Maintenance: a node state used before upgrades or disruptive host actions.
  • Recovery state: the current retry / cleanup / manual-attention state for an operation.
  • Template provenance: the recorded template version and inputs used to deploy a stack.

​
Expected result

Terms used across the dashboard and docs feel consistent instead of ambiguous.

What is StackShift

A precise description of StackShift as it exists today: a self-hosted application platform control plane for deploying and operating workloads across connected nodes.

How StackShift works

A product-level walkthrough of how the control plane, worker, agent, and runtime hosts cooperate.