deploy

name	deploy
description	Deploy the implemented feature to Azure once spec / plan / tasks / implement are complete.

User Input

$ARGUMENTS

Outline

$ARGUMENTS (if any) = deployment guidance: override the AZD environment from /azure.md, or --what-if for a dry run. Empty = "deploy to the AZD environment recorded in /azure.md".

1. Load context. Read specs/feature.json → <feature_directory>. Read spec.md, plan.md, tasks.md, tasks.json, and the workspace-root /azure.md. Stop if tasks.* is missing (run /speckit:tasks) or any task is still pending (run /speckit:implement; list the pending IDs). Stop immediately if /azure.md records none for the AZD template (no azure.yaml to deploy). Constitution non-negotiables are enforced upstream by /speckit:plan and /speckit:implement and are not re-checked here. Recommend running /speckit:verify first if it hasn't been run for the current implementation; proceed only with explicit user confirmation.

2. Tear down — only what you started. Offer to stop the local processes you started; never stop anything you didn't.

3. Confirm target. Show the AZD environment, subscription, resource group, and region from /azure.md and ASK the user to confirm before any state-changing command.

4. Prepare deployment. Use ACR cloud build for container images - When a service ships as a container image, build it with remoteBuild: true in azure.yaml, instead of a local docker build. Removes the local Docker dependency and keeps builds reproducible across machines and CI.

5. Deploy. Sign in if needed (azd auth login), then run:

   azd up -e <AZD environment> --debug > ./azure.log
   

   Stream output. On failure, halt and report the failing resource + error + likely next step (template fix, quota, RBAC); do not retry blindly. Never pass destructive flags (e.g. --force) without explicit user opt-in in $ARGUMENTS. Never print or commit secrets — values must come from Key Vault / managed identity.

6. Update status. Create or update <feature_directory>/deploy-status.json with the result of this deployment run. Overwrite the file atomically — do not append — and never write secrets (connection strings, keys, SAS tokens, passwords). Use ISO-8601 UTC timestamps (YYYY-MM-DDTHH:MM:SSZ). Use this exact schema:

   {
     "feature": "<feature_directory relative path>",
     "azd_environment": "<AZD environment>",
     "subscription_id": "<Azure Subscription Id>",
     "subscription_name": "<Azure Subscription Name>",
     "resource_group": "<resource group name>",
     "region": "<Azure region>",
     "azd_template": "<azd template repo or 'minimal'>",
     "status": "succeeded | failed | partial | what-if",
     "started_at": "2026-05-11T14:02:11Z",
     "completed_at": "2026-05-11T14:09:47Z",
     "duration_seconds": 456,
     "command": 
      
      
            
            
     
      
              
               
     
      
        
        
     
      
   
   

7. Document. Update documentation in /docs. If the azure-cost skill exists, use it to produce /docs/azure-cost-estimation.md with the cost estimate for the deployed environment.

8. Report. App URL/endpoint / Subscription / resource group / AZD environment, command run, resources created / updated / deleted, key outputs (IDs — no secrets), unresolved issues and follow-ups.