NEXUS AI CLI Deployment Guide

Most deployment guides hand you a twelve-step Docker workflow before you've written a line of application code. NEXUS AI CLI deployments skip that entirely — point the CLI at a GitHub repo, and NEXUS AI detects the framework, builds the container, and ships it. This guide covers the full lifecycle: from first install to automated production pipelines.

By Platform Super Admin • April 25, 2026 • Tutorial

Published: April 25, 2026
Category: Platform · DevOps
Reading time: 16 minutes
Author: NEXUS AI Team

Most deployment guides hand you a twelve-step Docker workflow before you've written a line of application code. NEXUS AI CLI deployments skip that entirely — point the CLI at a GitHub repo, and NEXUS AI detects the framework, builds the container, and ships it. This guide covers the full lifecycle: from first install to automated production pipelines.

By the end you'll know how to create, manage, scale, and roll back deployments using only the NEXUS AI CLI — no cloud console required.

What makes NEXUS AI CLI deployments different

Every major cloud has a CLI. What sets NEXUS AI apart is source-based deployment: you give the CLI a repo URL, and NEXUS AI handles everything downstream — framework detection, container builds, TLS provisioning, and autoscaling configuration.

The practical effect:

• No Dockerfile required. NEXUS AI detects Python, Node.js, Go, and other runtimes from your project root and builds a production image automatically.
• One command covers creation, secrets, and domains. You don't context-switch between five tools to get a service live.
• Multi-cloud with a single interface. The same CLI commands deploy to GCP Cloud Run, AWS ECS Fargate, or Azure Container Apps. Changing providers is a flag, not a migration.
• Full lifecycle support. Stop, start, scale, roll back, tail logs — all from the terminal you're already in.

Install the NEXUS AI CLI

Linux

curl -fsSL https://nexusai.run/install.sh | bash

The script detects your distro (Ubuntu/Debian, RHEL/CentOS, Alpine, Arch, openSUSE), installs Node.js 20 LTS if needed, and adds nexus to your PATH.

macOS

curl -fsSL https://nexusai.run/install-mac.sh | bash

Installs via Homebrew on Intel and Apple Silicon. Xcode Command Line Tools are required (the script prompts if missing).

Manual (any platform with Node.js 18+)

npm install -g nexusapp-cli

Verify the install:

nexus --version

Authenticate

nexus auth login

Opens a browser window to authenticate with your NEXUS AI account. After login, your token is stored in ~/.nexusai/config.json and used automatically by all subsequent commands.

Your first NEXUS AI CLI deployment

Step 1 — Source-based deployment (recommended)

Source-based deployment is the fastest path from code to production. No Docker, no image registry, no build pipeline to configure.

nexus deploy source \
  --repo https://github.com/your-org/my-api \
  --name my-api \
  --port 8000 \
  --provider gcp_cloud_run

What happens next:

1. NEXUS AI clones your repository
2. Detects the runtime (Python requirements.txt, Node.js package.json, Go go.mod, etc.)
3. Builds a production container image
4. Deploys to the specified provider with TLS and autoscaling enabled

Available providers: gcp_cloud_run, aws_ecs_fargate, azure_container_apps, docker (local).

Private repositories: pass a secret name containing your repo token:

nexus deploy source \
  --repo https://github.com/your-org/private-api \
  --name private-api \
  --port 3000 \
  --repo-secret GITHUB_TOKEN

Override detection: if auto-detection picks the wrong framework, you can specify explicitly:

nexus deploy source \
  --repo https://github.com/your-org/my-api \
  --name my-api \
  --port 8000 \
  --framework python \
  --build-command "pip install -r requirements.txt" \
  --start-command "uvicorn main:app --host 0.0.0.0 --port 8000"

Step 2 — Check deployment status

nexus deploy status my-api

Returns the current state (PENDING, BUILDING, RUNNING, FAILED), provider region, and the public URL once the deployment is live.

Step 3 — Tail logs

nexus deploy logs my-api --follow

Streams live logs from the running container. Drop --follow to get the last 100 lines and exit:

nexus deploy logs my-api --tail 50

Image-based deployment

If you already have a container image in a registry (Docker Hub, GHCR, ECR, GCR), deploy it directly:

nexus deploy create \
  --image your-org/my-api:latest \
  --name my-api \
  --port 8000 \
  --provider aws_ecs_fargate

Override the container's default startup command:

nexus deploy create \
  --image your-org/my-api:latest \
  --name my-api \
  --port 8000 \
  --start-command "node server.js --prod"

Image-based deployments and source-based deployments use the same management commands after creation — the path to production is the only difference.

Managing NEXUS AI CLI deployments

Secrets

Never put API keys in environment variables passed on the command line. Use NEXUS AI secrets instead — values are encrypted at rest and injected at runtime:

# Create a secret
nexus secret create OPENAI_API_KEY \
  --deployment my-api \
  --environment Production

# List secrets (values are hidden)
nexus secret list

# Update a secret value
nexus secret update <secret-id> --value sk-new-value

# Delete a secret
nexus secret delete <secret-id>

Custom domains

# Add a domain
nexus domain add api.yourcompany.com --deployment my-api

# List domains on a deployment
nexus domain list --deployment my-api

# Trigger DNS verification
nexus domain verify api.yourcompany.com --deployment my-api

NEXUS AI provisions TLS automatically after DNS verification. Point your CNAME at the provided record and the certificate is issued within minutes.

Scaling

# Scale to 3 replicas
nexus deploy scale my-api --replicas 3

# Scale back down
nexus deploy scale my-api --replicas 1

Replicas can be set between 1 and 10. GCP Cloud Run and Azure Container Apps also autoscale based on request load — the --replicas flag sets the minimum floor.

Health checks

nexus deploy health my-api

Returns the health status and recent health check logs. Useful for diagnosing why a deployment is stuck in PENDING or cycling between RUNNING and UNHEALTHY.

Stop and start

# Stop a deployment (preserves config, frees compute)
nexus deploy stop my-api

# Restart it later
nexus deploy start my-api

Stopping a deployment is non-destructive — all configuration, secrets, and domain mappings are preserved. Use it for dev/staging environments you don't need running 24/7.

Rollback

When a bad deploy makes it to production:

# Roll back to the previous version
nexus deploy rollback my-api

# Roll back to a specific revision
nexus deploy rollback my-api --target <previous-deployment-id>

Rollback redeploys the previous working configuration — same image or source, same environment, same secrets. No manual config reconstruction.

Automate NEXUS AI CLI deployments with GitHub Actions

Source-based CI/CD

The simplest pipeline: push to main, NEXUS AI rebuilds from source and redeploys.

# .github/workflows/deploy.yml
name: Deploy to NEXUS AI

on:
  push:
    branches: [main]

jobs:
  deploy:
    runs-on: ubuntu-latest
    steps:
      - name: Install NEXUS AI CLI
        run: curl -fsSL https://nexusai.run/install.sh | bash

      - name: Redeploy from source
        run: nexus deploy redeploy my-api
        env:
          NEXUSAI_TOKEN: ${{ secrets.NEXUSAI_TOKEN }}

nexus deploy redeploy pulls the latest commit from the repo URL that was used when the deployment was created, rebuilds the container, and does a rolling update.

Image-based CI/CD

For teams already building images in CI:

name: Build and Deploy

on:
  push:
    branches: [main]

jobs:
  deploy:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4

      - name: Build and push image
        run: |
          docker build -t ghcr.io/${{ github.repository }}:${{ github.sha }} .
          echo "${{ secrets.GITHUB_TOKEN }}" | docker login ghcr.io -u ${{ github.actor }} --password-stdin
          docker push ghcr.io/${{ github.repository }}:${{ github.sha }}

      - name: Install NEXUS AI CLI
        run: curl -fsSL https://nexusai.run/install.sh | bash

      - name: Deploy new image
        run: |
          nexus deploy redeploy my-api \
            --image ghcr.io/${{ github.repository }}:${{ github.sha }}
        env:
          NEXUSAI_TOKEN: ${{ secrets.NEXUSAI_TOKEN }}

Environment-specific pipelines

Deploy to staging on every push, production on tagged releases:

on:
  push:
    branches: [main]
    tags: ['v*']

jobs:
  deploy-staging:
    if: github.ref_type == 'branch'
    runs-on: ubuntu-latest
    steps:
      - run: nexus deploy redeploy my-api-staging
        env:
          NEXUSAI_TOKEN: ${{ secrets.NEXUSAI_TOKEN }}

  deploy-production:
    if: github.ref_type == 'tag'
    runs-on: ubuntu-latest
    steps:
      - run: nexus deploy redeploy my-api
        env:
          NEXUSAI_TOKEN: ${{ secrets.NEXUSAI_TOKEN }}

NEXUS AI CLI command reference

Frequently asked questions

Do I need Docker installed to use the NEXUS AI CLI?

No. For source-based deployments (nexus deploy source), NEXUS AI builds the container image on its own infrastructure. Docker is only needed if you're building images locally before deploying with nexus deploy create.

Which cloud providers does the NEXUS AI CLI support?

The CLI supports GCP Cloud Run (gcp_cloud_run), AWS ECS Fargate (aws_ecs_fargate), Azure Container Apps (azure_container_apps), and local Docker (docker). The provider flag is set at deployment creation and can be changed on redeploy.

How are secrets stored?

Secrets are encrypted at rest using AES-256-GCM and injected as environment variables at container start. Secret values are never returned by the API after creation — only the key name and metadata are visible via nexus secret list.

Can I use the NEXUS AI CLI in CI/CD without interactive login?

Yes. Set the NEXUSAI_TOKEN environment variable to your API token. The CLI checks this variable first and skips the browser-based login flow entirely. Generate a token in your NEXUS AI account settings.

What you can do with NEXUS AI CLI deployments

The NEXUS AI CLI covers the full lifecycle that normally spans four or five separate tools: a container registry, a cloud console, a secrets manager, a DNS panel, and a logging service. From a single terminal:

• Ship source code to production without writing a Dockerfile
• Manage encrypted secrets without leaving the CLI
• Roll back a bad deploy in one command
• Automate the entire pipeline with a 15-line GitHub Actions workflow

The NEXUS AI CLI installation guide has platform-specific setup instructions for Linux and macOS. Run nexus auth login and your first deployment is one command away.