How do I install the DanubeData CLI?

Run npm install -g @danubedata/cli. The CLI requires Node.js 18 or later. After installation, authenticate with danube auth to open browser-based login, or use danube login for token-based auth.

How do I authenticate?

The easiest way is danube auth — it opens your browser, you log in to DanubeData, click Authorize, and the CLI automatically receives a token. For CI/CD, use danube login --token or set the DANUBE_TOKEN environment variable.

Can I use the CLI in CI/CD pipelines?

Yes. Set DANUBE_TOKEN, DANUBE_TEAM_ID, and DANUBE_SITE_ID as environment variables. The CLI detects CI environments and skips interactive prompts automatically. Use the --no-wait flag for async deployments.

How do I manage S3 storage from the CLI?

Use danube storage buckets ls to list buckets, danube storage buckets create to create one, and danube storage keys create to generate S3 access keys. The secret key is shown once — save it immediately.

What files are included in a deployment?

The CLI packages all files in the deploy directory, respecting .gitignore patterns automatically. It always excludes .git, node_modules, and .danube directories. You can add extra ignore patterns in danube.json.

Can I roll back a deployment?

Yes. Use danube pages deployments ls to see all deployments with their revision numbers, then danube pages deployments rollback to activate a previous version instantly.

How do custom domains work?

Add a domain with danube pages domains add example.com. The CLI will show you the DNS record to create. Once configured, run danube pages domains verify example.com to trigger verification. TLS certificates are provisioned automatically.

Can an AI agent drive the CLI?

That is what it is built for. Every command accepts --json and returns one envelope: {success, data, error, meta}, on stdout, for failures as well as successes. Failures carry a machine-readable code and a retryable flag, exit codes are typed (2 means the command was wrong and is not worth retrying, 3 not authenticated, 4 not found), and --project scopes any invocation so nothing ever prompts. See docs.danubedata.ro/rapids-automation for the full machine contract.

How does an agent know when a deployment has finished?

Poll operation.terminal, never the status string — inferring terminality from text is how automated loops hang forever. danube operations wait blocks until terminal and honours the server-supplied poll interval, so pacing can change without updating your agent. If it times out you get operation.wait_timeout, which means the work has not finished rather than that it failed.

What stops a retry from creating duplicate resources?

Pass --idempotency-key to danube rapids apply. The same key replays the original result rather than creating a second container, so a network timeout followed by an automatic retry cannot leave duplicates behind. It is the failure mode that costs the most in unattended automation.

Is the CLI open source?

Yes. The CLI is published on npm as @danubedata/cli and the source code is available on GitHub. Contributions and issues are welcome.

Command Line Tool

DanubeData CLI

Name: DanubeData CLI
Brand: DanubeData
Availability: InStock

Manage your infrastructure from the terminal. Deploy sites, manage S3 storage, and authenticate with one command.

npm install

Node.js 18+

CI/CD Ready

Open Source

Get Started Free View on npm

terminal

$ npm install -g @danubedata/cli

$ danube auth
✔ Authenticated as Adrian

$ danube vps create --name web --image ubuntu-24.04
✔ Created VPS web (019d1234...)

$ danube vps ls
ID         NAME  STATUS   IP             PLAN
019d1234.. web   running  148.251.5.139  nano_shared

$ danube storage buckets create --name assets
✔ Created bucket assets

npm install -g @danubedata/cli

Customer stories

Don’t trust what we say — trust what our customers say.

“Danube Data offers easy and reliable access to S3 object instances. Everything is based in the EU. The web interface is simple to use and the API works without any problems. The Terraform provider is always up to date. I created a support ticket to report a bug and it was resolved the same day. I can fully recommend Danube Data and its services. If you have any questions, you can contact me via https://radiantgarden.io.”

Romario Muza

Founder at Radiant Garden e.K.

View all testimonials

The playground.

Every command, every flag — click one and watch it run. A live IDE-style preview of @danubedata/cli.

~/fleetVPS

danube vps ls

List all VPS instances

SYNOPSIS danube vps ls

$ npm install -g @danubedata/cli · Explore the danube CLI → ·Free & open source · MIT

14 Command Groups, One Install

Servers, databases, caches, storage, containers and registries — driven by hand or by an agent.

13 subcommands

danube vps

Full VPS lifecycle management. Create, start, stop, reboot, reinstall, resize, view metrics, and retrieve SSH passwords without leaving your shell.

~/fleet · 4 vps live

$ danube vps create --name web --image ubuntu-24.04

✓ created vps-019d1234 in 18s · ssh root@49.13.42.71

$ danube vps metrics web

cpu 12%

mem 1.2 / 8 GB

qps 1.8k

p95 3.4ms

danube auth

Authenticate via browser. Opens your browser, you log in and authorize, and the CLI receives a token automatically.

danube rapids

Serverless containers end to end: create, apply, deploy, scale, then logs, revisions, events, diagnose, preflight and probe when something needs explaining.

danube database

Managed PostgreSQL, MySQL and MariaDB. Create, resize, read replicas, snapshots, credentials, and both public and in-cluster hostnames.

danube cache

Redis, Valkey and Dragonfly. Create, resize, snapshot, and fetch connection details without opening the console.

danube registry

Private container registry discovery. Find the namespace you may push to, and verify a push would succeed before running it.

danube operations

Long-running work, addressable. Inspect an operation or block until it reaches a terminal state, with the server setting the poll interval.

danube storage buckets

Manage S3-compatible storage buckets. Create, update settings, set quotas, enable encryption, and view metrics.

danube storage keys

Manage S3 access keys. Create keys with secret display, list, view details, and revoke access.

danube pages deploy

Package your files and deploy to DanubeData. Respects .gitignore, polls build status, and shows your live URL.

danube pages domains

Add, verify, and remove custom domains. TLS certificates are provisioned automatically after DNS verification.

danube vps images

Browse available OS images grouped by distribution. Ubuntu, Debian, AlmaLinux, Rocky, Fedora, and Alpine.

danube whoami

Show the authenticated user, email, and all teams you belong to.

Real-World Examples

Copy and paste these commands to get started

# Install the CLI globally
$ npm install -g @danubedata/cli

# Authenticate via browser (like gh auth login)
$ danube auth
✔ Authenticated as Adrian (adrian@danubedata.ro)

# Create a VPS instance
$ danube vps create --name web-server --image ubuntu-24.04 --plan nano_shared
✔ Created VPS web-server (019d1002-c21a...)

# List your infrastructure
$ danube vps ls
$ danube storage buckets ls
$ danube pages deploy

Built for Developer Workflows

Everything you need for fast, reliable deployments

one envelope · every command

Built for Agentic Development

Every command accepts --json and answers with one envelope. Failures carry a code and a retryable flag, terminality is an explicit boolean, and idempotency keys make retries safe. An AI agent can deploy, verify and repair without a human reading a dashboard.

agent · --json no human

$ danube rapids apply --name api --wait --json

{ "success": true,

"data": { "action": "updated", "settled": true },

"error": null, "meta": { "count": 1 } }

on failure → "error": { "code", "retryable" } · exit 1–5

Browser-Based Auth

Run danube auth and log in via your browser. No manual token copying, no env-var dance — just like gh auth login.

VPS Management

Create, start, stop, reboot, and resize VPS instances. View metrics, retrieve passwords, and reinstall OS.

S3 Storage Management

Create buckets, manage access keys, set quotas, enable encryption, and view metrics from the terminal.

CI/CD Ready

Set environment variables and deploy headlessly. Detects CI environments and skips interactive prompts.

One-Command Deploy

Run danube pages deploy and your site is live. No configuration files required to get started.

Domain Management

Add, verify, and remove custom domains from your terminal. TLS certificates handled automatically.

Instant Rollbacks

Every deployment is versioned. Roll back to any previous revision with a single command.

Secure by Default

API tokens stored locally and never logged. Password generation and confirmation prompts for sensitive operations.

Fits Every Workflow

From solo developers to automated pipelines

headless · zero config

CI/CD Automation

Add deploy steps to GitHub Actions, GitLab CI, or any pipeline. Environment variables give you zero-config auth in headless runners.

GitHub ActionsGitLab CIAutomated Deploy

.github/workflows/deploy.yml actions

- run: npm install -g @danubedata/cli

- run: danube pages deploy --dir ./dist

env:

DANUBE_TOKEN: ${{ secrets.DANUBE_TOKEN }}

✓ deployed in 22s · live at site.pages.danubedata.ro

Server Fleet Management

Spin up VPS instances, check status, view metrics, and manage power state — all without leaving your terminal.

VPS CreateStart/StopMetrics

Storage & Backups

Create S3 buckets, generate access keys, set quotas, and manage encryption for your application data.

S3 BucketsAccess KeysQuotas

Team Collaboration

Share access across your team with API tokens. Every team member can manage servers, storage, and deployments.

Shared AccessTeam TokensAudit Trail

CLI FAQ

Authentication, scripts, scopes.

Run npm install -g @danubedata/cli. The CLI requires Node.js 18 or later. After installation, authenticate with danube auth to open browser-based login, or use danube login for token-based auth.
The easiest way is danube auth — it opens your browser, you log in to DanubeData, click Authorize, and the CLI automatically receives a token. For CI/CD, use danube login --token or set the DANUBE_TOKEN environment variable.
Yes. Set DANUBE_TOKEN, DANUBE_TEAM_ID, and DANUBE_SITE_ID as environment variables. The CLI detects CI environments and skips interactive prompts automatically. Use the --no-wait flag for async deployments.
Use danube storage buckets ls to list buckets, danube storage buckets create to create one, and danube storage keys create to generate S3 access keys. The secret key is shown once — save it immediately.
The CLI packages all files in the deploy directory, respecting .gitignore patterns automatically. It always excludes .git, node_modules, and .danube directories. You can add extra ignore patterns in danube.json.
Yes. Use danube pages deployments ls to see all deployments with their revision numbers, then danube pages deployments rollback <revision> to activate a previous version instantly.
Add a domain with danube pages domains add example.com. The CLI will show you the DNS record to create. Once configured, run danube pages domains verify example.com to trigger verification. TLS certificates are provisioned automatically.
That is what it is built for. Every command accepts --json and returns one envelope: {success, data, error, meta}, on stdout, for failures as well as successes. Failures carry a machine-readable code and a retryable flag, exit codes are typed (2 means the command was wrong and is not worth retrying, 3 not authenticated, 4 not found), and --project scopes any invocation so nothing ever prompts. See docs.danubedata.ro/rapids-automation for the full machine contract.
Poll operation.terminal, never the status string — inferring terminality from text is how automated loops hang forever. danube operations wait blocks until terminal and honours the server-supplied poll interval, so pacing can change without updating your agent. If it times out you get operation.wait_timeout, which means the work has not finished rather than that it failed.
Pass --idempotency-key to danube rapids apply. The same key replays the original result rather than creating a second container, so a network timeout followed by an automatic retry cannot leave duplicates behind. It is the failure mode that costs the most in unattended automation.
Yes. The CLI is published on npm as @danubedata/cli and the source code is available on GitHub. Contributions and issues are welcome.

Manage Everything from Your Terminal

Auth, deploy, manage storage. Your infrastructure at your fingertips.

npm install -g @danubedata/cli

Get Started Free View on npm

Get Started Free

Compute

Storage

Managed Apps

Caching

Databases

Messaging

Monitoring

DanubeData CLI

Automation & Agents

Tools

Developer Docs

Learn

Support

DanubeData CLI

Don’t trust what we say — trust what our customers say.

14 Command Groups, One Install

Real-World Examples

Built for Developer Workflows

Built for Agentic Development

Browser-Based Auth

VPS Management

S3 Storage Management

CI/CD Ready

One-Command Deploy

Domain Management

Instant Rollbacks

Secure by Default

Fits Every Workflow

CI/CD Automation

Server Fleet Management

Storage & Backups

Team Collaboration

Documentation & Resources

npm Registry

GitHub Repository

DanubeData Help

Explore More Solutions

Static Sites

Virtual Machines

Terraform

Object Storage

Manage Everything from Your Terminal