CLI Reference

The flui CLI is the primary surface for operating a Flui installation: creating the control cluster, managing workload clusters and their nodes, deploying applications, running backups, and switching between isolated installations on the same machine. The CLI and the dashboard overlap on most day-to-day operations, so once the platform is up either surface gets the job done; a handful of operations, though — creating or destroying the control cluster itself — live only here.

How the CLI is organised

Commands are grouped into topics — app, env, backup, context, and so on. Each topic owns a small set of verbs that act on the same kind of object: flui app build, flui env create, flui backup job run.

Every command exposes --help, which prints its usage, flags, and examples. When in doubt, flui <topic> --help lists the verbs available in that topic.

Topics

Topic	What it does	Reference
env	Lifecycle of the control cluster: create, inspect, scale, destroy.	Environments
cluster / node	Workload-cluster destroy and worker-node add/remove/list.	Clusters and nodes
app	Inspect, scale, restart, snapshot, redeploy, and delete applications; read logs, metrics, and crash reports.	Applications
deploy / catalog	Deploy from a flui.yaml manifest; validate the manifest offline.	Deploy
integration / repo	Connect the Flui GitHub App and import repositories into the installation — prerequisites for the first source-built deploy.	Integrations
template	Create a new repository from a Flui framework template.	Templates
backup	Manage backup destinations, policies, jobs, and restores.	flui backup --help (dedicated reference page coming soon)
auth	OIDC login and long-lived M2M API keys.	Authentication
context / config	Profiles (isolated installations) and layered configuration.	Profiles and config
server-types / ssh	List provider server sizes; SSH into a node.	Clusters and nodes
dev	Helpers for Flui contributors (export secrets, SSH tunnel).	Dev tools
dns / reconcile / update	Maintenance utilities.	Utilities

The full topic list is also available at the command line via flui --help.

Common conventions

A handful of flags recur across the CLI; once you know them, most commands behave consistently.

• -c, --cluster <name|id> — pick the target cluster explicitly on the commands that act on one. If the installation has a single workload cluster, it is auto-detected.
• -o, --output <format> — switch between human and machine-readable output. The exact set varies by command, but json is always available for scripting; the human format is text or table depending on the command’s shape.
• --no-wait / --detach — on the commands that submit a long-running operation, return as soon as the job is queued instead of blocking until it reaches a terminal status.
• -f, --force / -y, --yes — skip confirmation on destructive operations.
• --dry-run — available on a subset of commands that mutate remote state or local files; prints what would happen.

Commands that talk to the API print a short context banner before running (active profile + API URL) so it is always clear which installation the command is operating on.

Profiles on disk

The CLI keeps each installation’s state under ~/.flui/profiles/<name>/. A profile holds the API token, the encrypted credentials needed to talk to the provider, the SSH CA used for cluster nodes, and the local inventory of clusters/nodes/firewalls/operations the profile knows about. The path can be moved with the FLUI_DIR environment variable — see Environment variables.

Switching between installations is a flui context use <name> away; the rest of the CLI then targets the newly active profile transparently.