### Install Defang MCP Server Setup

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/concepts/mcp.md

Run this command to set up the Defang MCP Server for your chosen IDE client. Ensure the Defang CLI is installed first. Restart your IDE after setup.

```bash
defang mcp setup --client=<claude-code|claude-desktop|cursor|kiro|rovo|vscode|vscode-insiders|windsurf>
```

--------------------------------

### Install Dependencies

Source: https://github.com/defanglabs/defang-docs/blob/main/AGENTS.md

Use this command to install project dependencies.

```bash
npm install
```

--------------------------------

### Start Local Development Server

Source: https://github.com/defanglabs/defang-docs/blob/main/README.md

Starts a local development server and automatically opens the website in your browser. Useful for testing changes during development.

```bash
$ npm run start
```

--------------------------------

### Install Defang MCP Server via NPX

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/concepts/mcp.md

Alternative installation method using npx. This is supported but not recommended.

```bash
npx -y defang@latest mcp setup --client=<claude-code|claude-desktop|cursor|kiro|rovo|vscode|vscode-insiders|windsurf>
```

--------------------------------

### Install Defang CLI

Source: https://github.com/defanglabs/defang-docs/blob/main/blog/2025-10-10-n8n-aws-deployment.md

Installs the Defang CLI tool. Run this command in your terminal to get started.

```bash
eval "$(curl -fsSL s.defang.io/install)"
```

--------------------------------

### Install MCP Server for IDEs

Source: https://context7.com/defanglabs/defang-docs/llms.txt

Commands to install the MCP server for various IDEs. Use the specific command for your IDE or the npx alternative if preferred.

```bash
# Install MCP server for your IDE
defang mcp setup --client=vscode
defang mcp setup --client=cursor
defang mcp setup --client=claude-code
defang mcp setup --client=windsurf

# npx alternative (not recommended)
npx -y defang@latest mcp setup --client=vscode
```

--------------------------------

### Setup MCP Client for VS Code

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/concepts/mcp.md

Use this command to set up the Defang MCP client for VS Code. After setup, you can interact with the AI coding agent.

```bash
defang mcp setup --client=vscode
```

--------------------------------

### Example IDE Chat Prompts for MCP

Source: https://context7.com/defanglabs/defang-docs/llms.txt

Example prompts to use with AI coding assistants after setting up the MCP server. These demonstrate how to interact with Defang services via chat.

```text
# Example IDE chat prompts after MCP setup:
"can you deploy this to defang?"
"what services do I have?"
"please destroy this project."
"can you estimate balance mode on aws for this project?"
"set the defang config DATABASE_URL to example.com"
"list all the defang configs I have set for this project"
```

--------------------------------

### Install Defang CLI

Source: https://context7.com/defanglabs/defang-docs/llms.txt

Install the Defang CLI using a shell script, Homebrew, Winget, or Nix.

```bash
eval "$(curl -fsSL s.defang.io/install)"
```

```bash
brew install DefangLabs/defang/defang
```

```bash
winget install defang
```

```bash
nix profile install github:DefangLabs/defang#defang-bin --refresh
```

--------------------------------

### Example Cost Estimation Output

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/concepts/estimation.md

This example shows the estimated monthly cost for deploying the django-postgres sample to AWS using the balanced deployment mode. It includes a breakdown of costs per service and a total estimated monthly cost.

```bash
defang estimate --provider=aws --mode=balanced
 * Packaging the project files for django at /Users/defang/samples/samples/django-postgres/app
 * Generating deployment preview
 * Preparing estimate

Estimate for Deployment Mode: BALANCED

This mode strikes a balance between cost and availability. Your application
will be deployed with spot instances. Databases will be provisioned using
resources optimized for production. Services in the "internal" network will
be deployed to a private subnet with a NAT gateway for outbound internet access.

Cost    Quantity           Service   Description
$16.43  730 Hours          (shared)  AWSELB USW2-LoadBalancerUsage
$32.85  730 Hours          (shared)  AmazonEC2 USW2-NatGateway-Hours
$25.00  100 %Utilized/mo   db        AmazonRDS USW2-InstanceUsage:db.r5.large
$1.62   14600 GB-Hours     django    AmazonECS USW2-Fargate-EphemeralStorage-GB-Hours (20 GB * 730 hours)
$1.62   365 GB-Hours       django    AmazonECS USW2-Fargate-GB-Hours (0.50 GB * 730 hours)
-$1.14  365 GB-Hours       django    AmazonECS USW2-Fargate-GB-Hours-SpotDiscount (Estimated @ 70%)
$7.39   182.50 vCPU-Hours  django    AmazonECS USW2-Fargate-vCPU-Hours:perCPU (0.25 vCPU * 730 hours)
-$5.17  182.50 vCPU-Hours  django    AmazonECS USW2-Fargate-vCPU-Hours:perCPU-SpotDiscount (Estimated @ 70%)
Estimated Monthly Cost: $78.60 (+ usage)

Estimate does not include taxes or Discount Programs.
To estimate other modes, use defang estimate --mode=affordable|balanced|high_availability
For help with warnings, check our FAQ at https://s.defang.io/warnings

```

--------------------------------

### Deploy Project

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/tutorials/agent-skills.md

Navigate to your project directory and execute this command to deploy. The agent will guide you through the deployment steps.

```shell
/defang:deploy
```

--------------------------------

### Initialize Heroku Migration

Source: https://github.com/defanglabs/defang-docs/blob/main/blog/2025-09-25-heroku-migration.md

Run this command after installing and logging into the Defang CLI to start the migration process from Heroku. It prompts you to select the 'Migrate from Heroku' option.

```bash
defang init
```

--------------------------------

### Setup Defang MCP Server for Kiro

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/concepts/mcp.md

Use this command to set up the Defang MCP Server specifically for the Kiro IDE. After setup, interact with the AI agent using Defang actions.

```bash
defang mcp setup --client=kiro
```

--------------------------------

### Install Defang CLI with Homebrew

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/intro/getting-started.mdx

Installs the Defang CLI using Homebrew. Ensure you have Homebrew installed.

```bash
brew install DefangLabs/defang/defang
```

--------------------------------

### Setup MCP Client for Windsurf

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/concepts/mcp.md

Use this command to set up the Defang MCP client for Windsurf. After setup, you can interact with the AI coding agent.

```bash
defang mcp setup --client=windsurf
```

--------------------------------

### Non-CI/CD Stack Configuration Example

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/tutorials/deploying-from-github-actions/to-aws.mdx

This example shows a `.defang/` stack configuration that is NOT compatible with CI/CD due to the inclusion of `AWS_PROFILE`. Remove the `AWS_PROFILE` line for deployments from GitHub Actions.

```bash
# .defang/production - NOT CI/CD compatible (will fail)
AWS_PROFILE="my-local-profile"  # Remove this line
AWS_REGION="us-west-2"
DEFANG_PROVIDER="aws"
```

--------------------------------

### Install Project Dependencies

Source: https://github.com/defanglabs/defang-docs/blob/main/README.md

Installs all necessary Node.js dependencies for the project. Ensure you have cloned the required repositories in sibling directories.

```bash
$ npm install
```

--------------------------------

### Setup Defang MCP Server for Claude Desktop

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/concepts/mcp.md

Use this command to set up the Defang MCP Server specifically for the Claude Desktop client. After setup, interact with the AI agent using Defang actions.

```bash
defang mcp setup --client=claude-desktop
```

--------------------------------

### Setup Defang MCP Server for Cursor

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/concepts/mcp.md

Use this command to set up the Defang MCP Server specifically for the Cursor IDE. After setup, interact with the AI agent using Defang actions.

```bash
defang mcp setup --client=cursor
```

--------------------------------

### Example GitHub New Repository URL

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/tutorials/adding-custom-one-click-deploy.md

This is an example of the URL generated when creating a new repository from a template. It includes parameters for the template name and owner.

```text
https://github.com/new?template_name=<your-repo-name>&template_owner=<your-github-username>
```

--------------------------------

### Initialize Defang for Heroku Migration

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/tutorials/migrating-from-heroku.md

Start the Defang initialization process and select the 'Migrate from Heroku' option.

```bash
$ defang init
? How would you like to start?  [Use arrows to move, type to filter]
  Generate with AI
  Clone a sample
> Migrate from Heroku
```

--------------------------------

### Start Local Development Server

Source: https://github.com/defanglabs/defang-docs/blob/main/AGENTS.md

Runs the local development server with hot-reloading. This command also executes the prebuild script.

```bash
npm run start
```

--------------------------------

### Example Managed MongoDB Configuration

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/concepts/managed-storage/managed-mongodb.md

This example shows how to define a managed MongoDB service using the `x-defang-mongodb` extension in your `compose.yaml`. It includes environment variables for the database name and connection string.

```yaml
services:
  app:
    # [...]
    environment:
      DB_URL: mongodb://$MONGO_INITDB_ROOT_USERNAME:$MONGO_INITDB_ROOT_PASSWORD@db:27017/
    depends_on:
      - db

  db:
    image: mongo:5
    x-defang-mongodb: true
    environment:
      MONGO_INITDB_DATABASE: mydb
```

--------------------------------

### Install Defang CLI with Nix

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/intro/getting-started.mdx

Installs the Defang CLI using Nix. This command ensures your Nix profile is up-to-date before installation.

```bash
nix profile install github:DefangLabs/defang#defang-bin --refresh
```

--------------------------------

### Install Defang CLI with Winget

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/intro/getting-started.mdx

Installs the Defang CLI on Windows using Winget. This is a convenient option for Windows users.

```powershell
winget install defang
```

--------------------------------

### Setup MCP Client for VS Code Insiders

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/concepts/mcp.md

Use this command to set up the Defang MCP client for VS Code Insiders. After setup, you can interact with the AI coding agent.

```bash
defang mcp setup --client=vscode-insiders
```

--------------------------------

### Setup Defang MCP Server for Rovo

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/concepts/mcp.md

Use this command to set up the Defang MCP Server specifically for the Rovo platform. After setup, interact with the AI agent using Defang actions.

```bash
defang mcp setup --client=rovo
```

--------------------------------

### List Running Defang Services

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/tutorials/using-codespaces-gitpod.md

After successfully logging in, use this command to verify your Defang setup by listing any currently running services. This confirms that Defang is operational in your environment.

```bash
defang ls
```

--------------------------------

### Heroku Add-on Information

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/tutorials/migrating-from-heroku.md

Example output listing a PostgreSQL database add-on for a Heroku application.

```bash
heroku addons -a vast-badlands-production

 Add-on                                     Plan        Price        Max price State
 ────────────────────────────────────────── ─────────── ──────────── ───────── ───────
 heroku-postgresql (postgresql-rosy-12345) essential-0 ~$0.007/hour $5/month  created
  └─ as DATABASE
```

--------------------------------

### Defang Deployment Output

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/tutorials/deploy-to-playground.mdx

This is an example of the output you will see during and after a successful deployment. It includes build context upload, service deployment status, a link to monitor services, and the final URL where your service will be available.

```bash
* Uploading build context for app
* Deploying service app
* Monitor your services' status in the defang portal
  - https://portal.defang.dev/service/app
* Tailing logs for deployment ID o59k89vk3qc8 ; press Ctrl+C to detach:
* Press V to toggle verbose mode
2024-09-19T10:50:53.572443-07:00 cd Update started for stack jordanstephens-prod1
2024-09-19T10:51:05.536299-07:00 cd Update succeeded in 11.99769745s ; provisioning...
2024-09-19T10:51:39.419693-07:00 app Server running at http://0.0.0.0:3000/
* Service app is in state DEPLOYMENT_COMPLETED and will be available at:
  - https://jordanstephens-app--3000.prod1.defang.dev
* Done.
```

--------------------------------

### Initial Compose File

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/tutorials/deploy-openai-apps/aws-bedrock.mdx

This is the starting point for a Docker Compose file with a single application service.

```yaml
services:
  app:
    build:
      context: .
    ports:
      - 3000:3000
    environment:
      OPENAI_API_KEY:
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:3000/"]
```

--------------------------------

### Install Defang Pulumi Provider with npm

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/concepts/pulumi.md

Use this command to add the Defang Pulumi provider to your project when using npm.

```bash
npm i @defang-io/pulumi-defang
```

--------------------------------

### Setup Defang MCP Server for Claude Code

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/concepts/mcp.md

Use this command to set up the Defang MCP Server specifically for the Claude Code client. After setup, interact with the AI agent using Defang actions.

```bash
defang mcp setup --client=claude-code
```

--------------------------------

### Install Defang Pulumi Provider with pnpm

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/concepts/pulumi.md

Use this command to add the Defang Pulumi provider to your project when using pnpm.

```bash
pnpm i @defang-io/pulumi-defang
```

--------------------------------

### Clone Repositories

Source: https://github.com/defanglabs/defang-docs/blob/main/README.md

Clones the defanglabs/defang and defanglabs/samples repositories into sibling directories. This is a prerequisite for installation.

```bash
$ git clone https://github.com/DefangLabs/defang.git ../defang 
$ git clone https://github.com/DefangLabs/samples.git ../samples
```

--------------------------------

### Heroku Dyno Information

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/tutorials/migrating-from-heroku.md

Example output showing a single 'web' dyno for a Heroku application.

```bash
$ heroku ps -a vast-badlands-production

=== web (Eco): bin/rails server -p ${PORT:-5000} -e $RAILS_ENV (1)
```

--------------------------------

### Tail Live Logs for All Services

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/tutorials/monitoring-your-services.md

Use this command to view real-time logs from all your deployed services. Ensure the Defang CLI is installed and configured.

```bash
$ defang tail
```

--------------------------------

### Build and Run Docker Compose Locally

Source: https://github.com/defanglabs/defang-docs/blob/main/blog/2025-02-18-model-context-protocol.md

Use this command to build your Docker images and start your application services locally.

```bash
docker compose up --build
```

--------------------------------

### Complete Example Compose File for AWS Bedrock

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/tutorials/deploy-openai-apps/aws-bedrock.mdx

This is the complete Docker Compose file configuration for deploying an OpenAI app to AWS Bedrock using Defang's openai-access-gateway.

```yaml
services:
  app:
    build:
      context: .
    ports:
      - 3000:3000
    environment:
      OPENAI_API_KEY:
      OPENAI_BASE_URL: "http://llm/api/v1"
      MODEL: "anthropic.claude-3-sonnet-20240229-v1:0"
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:3000/"]

  llm:
    image: defangio/openai-access-gateway
    x-defang-llm: true
    ports:
      - target: 80
        published: 80
        mode: host
    environment:
      - OPENAI_API_KEY
      - REGION
```

--------------------------------

### Clone Sample Project

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/tutorials/estimating-gcp-deployment-costs.md

Clone a sample project and navigate into its directory to prepare for cost estimation.

```bash
defang generate       # select `django-postgres` (or any other sample) from the list
cd django-postgres    # or the name of your project's working directory
```

--------------------------------

### Run Defang CLI without Global Installation

Source: https://github.com/defanglabs/defang-docs/blob/main/blog/2024-05-01-may-product-updates.md

Use npx to execute the latest Defang CLI version without needing to install it globally. This simplifies getting started and ensures you're always using the most recent release.

```bash
npx defang@latest
```

--------------------------------

### Manual MCP Server Configuration

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/concepts/mcp.md

This JSON configuration can be manually added to your IDE's MCP file to set up the Defang MCP Server without using the setup command or installing the Defang CLI.

```json
{
  "mcpServers": {
    "defang": {
      "command": "npx",
      "args": ["-y", "defang@latest", "mcp", "serve"]
    }
  }
}
```

--------------------------------

### Navigate to Project Directory

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/tutorials/migrating-from-heroku.md

Change to your project's working directory before running `defang init`.

```bash
$ cd ~/w/vast-badlands
$ ls
Gemfile		Procfile	Rakefile	bin	config.ru	lib	public	test	vendor
Gemfile.lock	README.md	app	config	db	log	storage	tmp
```

--------------------------------

### Verify Defang CLI Installation

Source: https://github.com/defanglabs/defang-docs/blob/main/blog/2025-10-10-n8n-aws-deployment.md

Checks if the Defang CLI has been installed correctly. Use this after installation to confirm.

```bash
defang version
```

--------------------------------

### Build and Deploy with `defang compose up`

Source: https://context7.com/defanglabs/defang-docs/llms.txt

Build and deploy applications defined in `compose.yaml`. This command uploads source code, builds images in-cloud, provisions resources, and tails logs. Specify the cloud provider using `--provider` or the `DEFANG_PROVIDER` environment variable. Deployment modes like `high_availability` can also be set.

```bash
# Deploy to Defang Playground (default, non-production)
defang compose up

# Deploy to AWS BYOC
defang compose up --provider=aws

# Deploy to GCP
defang compose up --provider=gcp

# Deploy with a specific deployment mode
defang compose up --provider=aws --mode=high_availability

# Using environment variable instead of flag
export DEFANG_PROVIDER=digitalocean
defang compose up

# Expected output:
# * Uploading build context for app
# * Deploying service app
# * Tailing logs for deployment ID o59k89vk3qc8 ; press Ctrl+C to detach:
# 2024-09-19T10:51:39 app Server running at http://0.0.0.0:3000/
# * Service app is in state DEPLOYMENT_COMPLETED and will be available at:
#   - https://alice-app--3000.prod1.defang.dev
```

--------------------------------

### Install Defang Pulumi Provider

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/concepts/pulumi.md

Install the Defang Pulumi provider in your Pulumi project using npm, pnpm, or yarn.

```bash
npm i @defang-io/pulumi-defang

pnpm i @defang-io/pulumi-defang

yarn add @defang-io/pulumi-defang
```

--------------------------------

### Serve Production Build

Source: https://github.com/defanglabs/defang-docs/blob/main/AGENTS.md

Tests the generated production build by serving it locally.

```bash
npm run serve
```

--------------------------------

### Configure Managed Postgres with Defang

Source: https://context7.com/defanglabs/defang-docs/llms.txt

Demonstrates setting up a managed PostgreSQL instance using Defang's `x-defang-postgres` extension. Requires setting the password and SSL mode via `defang config set` before deployment.

```bash
# Must set the password before deploying
defang config set POSTGRES_PASSWORD=mysecretpassword
defang config set SSL_MODE=require
```

--------------------------------

### Defang Compose Deployment Modes

Source: https://context7.com/defanglabs/defang-docs/llms.txt

Demonstrates how to deploy applications using different Defang compose modes: affordable (development/test), balanced (staging), and high_availability (production). Each mode affects cost, logging, and update strategies.

```bash
# Development — spot instances, 1-day logs, no rolling updates
defang compose up --provider=aws --mode=affordable

# Staging — 7-day logs, single NAT gateway
defang compose up --provider=aws --mode=balanced

# Production — on-demand instances, 30-day logs, rolling updates, multiple NAT gateways
defang compose up --provider=aws --mode=high_availability
```

--------------------------------

### Generate HTTP GET Endpoint for Memes

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/concepts/generate.md

Use this prompt to generate an HTTP GET endpoint that returns an HTML page with a random meme from the imgflip API.

```text
An HTTP GET endpoint that returns an HTML with a random meme from the imgflip API
```

--------------------------------

### Deploy Application

Source: https://github.com/defanglabs/defang-docs/blob/main/blog/2025-09-25-heroku-migration.md

Execute this command to provision AWS infrastructure (ECS, RDS, VPC, etc.) and deploy your application. Defang handles database readiness and service startup.

```bash
defang up
```

--------------------------------

### Build for Production

Source: https://github.com/defanglabs/defang-docs/blob/main/AGENTS.md

Generates production-ready static output for the documentation site.

```bash
npm run build
```

--------------------------------

### Deploy Project to DigitalOcean

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/providers/digitalocean.md

Use the `--provider=digitalocean` flag or set the `DEFANG_PROVIDER` environment variable to deploy your project using the DigitalOcean provider.

```bash
$ defang compose up --provider=digitalocean
```

```bash
$ export DEFANG_PROVIDER=digitalocean
```

--------------------------------

### Defang Railpack Deployment Example

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/concepts/railpack.md

This Compose file example demonstrates how to configure a service for a Defang Railpack deployment. Ensure your Compose file does not explicitly mention a Dockerfile when using Railpack for automatic image building.

```yaml
services:
  app:
    restart: unless-stopped
    build:
      context: .
    deploy:
      resources:
        reservations:
          cpus: "1.0"
          memory: 512M
    ports:
      - mode: ingress
        target: 3000
        published: 3000
```

--------------------------------

### Navigate to Project Directory

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/tutorials/generate-new-code-using-ai.mdx

After the project is generated, change into the newly created project directory to review or modify the files.

```shell
cd project1
```

--------------------------------

### Example Managed Postgres Configuration

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/concepts/managed-storage/managed-postgres.mdx

An example of configuring application and database services for managed Postgres. Leaving `POSTGRES_PASSWORD` empty instructs Defang to use the value set via `defang config set POSTGRES_PASSWORD`. A connection string is constructed using interpolation.

```yaml
app:
  # [...]
  environment:
    POSTGRES_HOST: database
    POSTGRES_USER: postgres
    POSTGRES_DB: postgres
    // highlight-start
    # Note: by leaving the value empty, Defang will use the
    # value set using `defang config set POSTGRES_PASSWORD`
    POSTGRES_PASSWORD:
    // highlight-end
    # Note: you can create a connection string by using interpolation,
    # reference config variables by using ${<config name>}
    CONNECTURL: postgresql://postgres:${POSTGRES_PASSWORD}@database:5432/postgres?sslmode=require
database:
  image: postgres:18
  x-defang-postgres: true
  ports:
    - mode: host
      target: 5432
  environment:
    // highlight-start
    # Note: by leaving the value empty, Defang will use the
    # value set using `defang config set POSTGRES_PASSWORD`
    POSTGRES_PASSWORD:
    // highlight-end

```

--------------------------------

### Deploy Project to Cloud

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/concepts/estimation.md

After estimating costs, use this command to deploy your project. It supports multiple cloud providers (aws, azure, gcp, digitalocean) and deployment modes.

```bash
defang compose up [--provider aws|azure|gcp|digitalocean] [--mode affordable|balanced|high_availability]
```

--------------------------------

### Generate Project with AI

Source: https://context7.com/defanglabs/defang-docs/llms.txt

Generate a project skeleton from a natural language description or a named sample using the AI agent. After generation, navigate to the project directory and run `defang compose up` to build and deploy.

```bash
# Generate from a named sample
defang generate nodejs-http
# Prompts: "What folder would you like to create the project in?"
# → Writes: compose.yaml, Dockerfile, app/main.js, .github/workflows/deploy.yaml

# Generate from a natural language prompt
defang generate
# Select language: Nodejs
# Choose: "Generate with AI"
# Prompt: "A basic service with 2 REST endpoints. The default endpoint returns
#          { "status": "OK" } and /echo echoes request parameters."
# Folder: my-service
# → Generates all required project files

cd my-service
defang compose up
```

--------------------------------

### Set DigitalOcean Credentials and Deploy

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/tutorials/migrating-from-heroku.md

Set environment variables for DigitalOcean access and then deploy your application using Defang. Ensure your DigitalOcean token and Spaces credentials are correctly configured.

```bash
export DIGITALOCEAN_TOKEN=your_digitalocean_token
export SPACES_ACCESS_KEY_ID=your_spaces_access_key_id
export SPACES_SECRET_ACCESS_KEY=your_spaces_secret_access_key
$ defang compose up --provider digitalocean
```

--------------------------------

### Configure One-off Job with Dependencies

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/concepts/one-off-jobs.md

Set up a 'migrate' one-off job to run database migrations using './migrate.sh' only after the 'db' service is ready. The 'web' service then depends on the 'migrate' job.

```yaml
services:
  db:
    image: postgres:latest

  migrate:
    build: .
    command: ["./migrate.sh"]
    depends_on:
      - db
    restart: "no"

  web:
    build: .
    ports:
      - "80:80"
    depends_on:
      - migrate
```

--------------------------------

### Initialize Heroku Migration (Alternative)

Source: https://github.com/defanglabs/defang-docs/blob/main/blog/2025-09-25-heroku-migration.md

An alternative command to directly initiate the Heroku migration workflow, potentially skipping interactive prompts.

```bash
defang init migrate-from-heroku
```

--------------------------------

### Valid Healthcheck for Ingress Ports

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/intro/faq/warnings-errors.md

This is an example of a valid healthcheck configuration for services with ingress ports, which includes 'localhost'.

```yaml
healthcheck:
  test: ["CMD", "curl", "-f", "http://localhost:1234/health"]
```

--------------------------------

### Initialize Pulumi TypeScript Project

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/tutorials/deploy-using-pulumi.mdx

Create a new Pulumi project using the TypeScript template. The `--force` flag is used because the directory is not empty.

```bash
pulumi new typescript -y --force
```

--------------------------------

### Authenticate GCP CLI

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/tutorials/setting_up_your_gcp_account.md

Run this command after installing the gcloud CLI to generate application default credentials. These are used during deployment.

```bash
gcloud auth application-default login
```

--------------------------------

### Navigate to Project Directory

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/intro/getting-started.mdx

Changes the current directory to the newly generated project folder.

```bash
cd welcome-to-defang
```

--------------------------------

### Install Defang Pulumi Provider with yarn

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/concepts/pulumi.md

Use this command to add the Defang Pulumi provider to your project when using yarn.

```bash
yarn add @defang-io/pulumi-defang
```

--------------------------------

### Deploy Project to Azure

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/providers/azure.md

Deploy your project to Azure using the Defang CLI. Specify the Azure provider explicitly or set it as an environment variable.

```bash
defang compose up --provider=azure
```

```bash
export DEFANG_PROVIDER=azure
```

--------------------------------

### Configure Service Healthchecks (wget)

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/concepts/compose.md

Define healthcheck configurations for services, required for exposed ports. This example uses `wget`.

```yaml
healthcheck:
  test: ["CMD", "wget", "--spider", "http://localhost:8080/"]
  interval: 30s
  timeout: 90s
  retries: 3
```

--------------------------------

### Clone Slackbot Sample Code

Source: https://github.com/defanglabs/defang-docs/blob/main/blog/2024-03-28-slackbot-sample.md

Clone the Defang repository and navigate to the slackbot sample directory. This sets up the necessary files for deployment.

```bash
git clone https://github.com/DefangLabs/defang
cd defang/samples/golang/slackbot
```

--------------------------------

### Create New n8n Project Directory

Source: https://github.com/defanglabs/defang-docs/blob/main/blog/2025-10-10-n8n-aws-deployment.md

Creates a new directory for your n8n deployment and navigates into it. This is the first step in setting up your n8n project with Defang.

```bash
defang new n8n
cd n8n
```

--------------------------------

### Configure Service Healthchecks (curl)

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/concepts/compose.md

Define healthcheck configurations for services, required for exposed ports. This example uses `curl`.

```yaml
healthcheck:
  test: ["CMD", "curl", "-f", "http://localhost:8080/"]
  interval: 30s
  timeout: 90s
  retries: 3
```

--------------------------------

### Minimal Defang Service with Pulumi

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/concepts/pulumi.md

This is a minimal example of a Pulumi program that defines a Defang service. It requires the `@defang-io/pulumi-defang` package to be imported.

```typescript
import * as defang from "@defang-io/pulumi-defang/lib";

const service = new defang.DefangService("my-service", {
    image: "strm/helloworld-http:latest",
    ports: [{
        target: 80,
        mode: "ingress",
        protocol: "http",
    }],
});
```

--------------------------------

### Authenticate with AWS CLI

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/tutorials/setting_up_your_aws_account.mdx

Run this command after installing the AWS CLI to authenticate your local machine. It creates a profile with short-lived credentials for Defang.

```bash
aws login
```

--------------------------------

### List All Defang Stacks in a Project

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/concepts/stacks.md

View a list of all created stacks for the current project, including their provider, region, and mode.

```bash
defang stack list
```

--------------------------------

### Add Defang Skill using Codex

Source: https://github.com/defanglabs/defang-docs/blob/main/docs/tutorials/agent-skills.md

For Codex, manually add the Defang marketplace entry to your personal marketplace file. Then, install the plugin.

```json
{
  "name": "defang-skills",
  "owner": {
    "name": "DefangLabs"
  },
  "plugins": [
    {
      "name": "defang",
      "source": {
        "source": "github",
        "repo": "DefangLabs/defang",
        "path": "codex-plugin"
      },
      "category": "Infrastructure"
    }
  ]
}
```

```shell
/plugin install defang@defang-skills
```