Workflows

This page describes common multi-command workflows for the EAI CLI. Each workflow combines several commands into an end-to-end process for a specific task.

Set Up a New Vertical

Create a new vertical application from scratch and get it running locally.

Steps

# 1. Scaffold a new project from the template
eai init my-vertical

# 2. Move into the project directory
cd my-vertical

# 3. Authenticate with the platform
eai login

# 4. Pull environment configuration from the cloud
eai env pull

# 5. Validate your Object Type definitions
eai types validate

# 6. Seed Object Types to the platform
eai types seed

# 7. Start the local development server
eai dev

What each step does

Step	Command	Purpose
1	eai init	Clones the Vertical Template, installs dependencies, and sets up the project structure.
2	cd my-vertical	Navigate into the newly created project directory.
3	eai login	Opens a browser window for Microsoft Entra ID authentication and stores tokens locally.
4	eai env pull	Downloads API endpoints, tenant configuration, and other environment variables to .env.local.
5	eai types validate	Checks that the Object Type definitions in src/eai.config/object-types.ts are valid.
6	eai types seed	Pushes the validated Object Types to the platform so resources can be created.
7	eai dev	Starts the Next.js development server with platform connectivity on port 3000.

---

Manage Object Types

Define, validate, and sync your data model between local definitions and the platform.

Validate and seed

After editing src/eai.config/object-types.ts, validate and push your changes:

# Validate locally, then seed to the platform
eai types validate && eai types seed

Preview changes before seeding

Use diff and dry-run to review what will change:

# See differences between local and remote types
eai types diff

# Preview what would be seeded without making changes
eai types seed --dry-run

# If everything looks good, seed for real
eai types seed

Sync changes from the Configurator UI

If someone made changes through the web-based Configurator, pull them to your local definitions:

# Pull remote types to local TypeScript definitions
eai types pull

# Verify the pulled types are valid
eai types validate

Seed a specific tenant or environment

# Seed only a specific tenant
eai types seed --tenant-key my-tenant

# Seed to the staging environment
eai types seed --env staging

Full Object Type lifecycle

# 1. Validate definitions
eai types validate

# 2. Preview changes
eai types diff

# 3. Seed to development
eai types seed --dry-run
eai types seed

# 4. Verify schema is published
eai resources schema

Step	Command	Purpose
1	eai types validate	Catch schema errors before pushing to the platform.
2	eai types diff	Review additions, removals, and modifications.
3	eai types seed	Push Object Types to the platform (use --dry-run first).
4	eai resources schema	Confirm the types are published and available for resource operations.

---

Deploy Your Vertical

Set up the deployment pipeline and deploy your vertical application to Azure App Service.

First-time setup

# 1. Configure GitHub Actions and repository secrets
eai deploy setup --repo eai-tools/my-vertical

# 2. Trigger the first deployment
eai deploy trigger

# 3. Monitor the deployment
eai deploy status

Subsequent deployments

After the initial setup, you only need to trigger and monitor:

# Trigger a new deployment
eai deploy trigger

# Check deployment status
eai deploy status

Pre-deployment checklist

Before deploying, verify that everything is in order:

# 1. Run the doctor to check for issues
eai doctor

# 2. Verify platform connectivity
eai verify

# 3. Validate Object Types
eai types validate

# 4. Deploy
eai deploy trigger

Step	Command	Purpose
1	eai deploy setup	Generates the GitHub Actions workflow and configures secrets. Only needed once.
2	eai deploy trigger	Kicks off the build and deployment to Azure.
3	eai deploy status	Shows build progress, deployment state, and any errors.

---

Set Up a New Tenant

Create a new tenant and configure its environment.

Steps

# 1. Create the tenant
eai tenant create \
  --name "My Application" \
  --slug my-app \
  --domain "myapp.com,app.myenterprise.ai"

# 2. Push environment configuration to the new tenant
eai env push

# 3. Seed Object Types for the tenant
eai types seed --tenant-key my-app

# 4. Verify the tenant is operational
eai verify

Step	Command	Purpose
1	eai tenant create	Creates a new isolated tenant with the specified name, slug, and domains.
2	eai env push	Pushes environment configuration so the tenant has the correct API settings.
3	eai types seed	Seeds Object Types for the specific tenant so resources can be created.
4	eai verify	Confirms that all platform services are accessible for the new tenant.

---

Troubleshooting Workflow

When something is not working, follow this diagnostic sequence:

# 1. Run comprehensive diagnostics
eai doctor

# 2. Check platform connectivity
eai verify

# 3. Check your authentication status
eai whoami

# 4. If issues are found, try automatic fixes
eai doctor --fix

# 5. If auth is expired, re-authenticate
eai logout && eai login

# 6. Re-sync environment configuration
eai env pull