URA Deploy Skill

Deploy a new version of Universal Room Automation. Runs the full pipeline: tests → README → deploy.sh → HACS download → config check → HA restart → verify.

Usage

/deploy <version> "<summary>" "<release-notes>"

Example: /deploy 3.6.0.12 "Zone camera discovery" "Added camera-to-zone mapping via area registry"

Steps

When invoked, execute these steps IN ORDER. Stop on any failure.

Step 1: Run Tests

bash
1PYTHONPATH=quality python3 -m pytest quality/tests/ -v

If any tests fail, STOP and report the failures. Do not proceed.

Step 2: Verify README Exists

Check that docs/readmes/README_v<version>.md exists. If not, create it based on the changes in git diff.

Step 3: Pre-stage New Files

Run git status to find any new directories or files that deploy.sh might miss (it only globs *.py in the component dir, not subdirectories). Stage them with git add.

Step 4: Deploy

bash
1./scripts/deploy.sh <version> "<summary>" "<release-notes>"

Report the PR URL and release URL from the output.

Step 5: HACS Update

Use the mcp__home-assistant__ha_hacs_download tool:

• repository_id: "ojiudezue/universal-room-automation"
• version: "v<version>"

Step 6: Config Check

Use mcp__home-assistant__ha_check_config to validate HA configuration. If invalid, STOP and report errors.

Step 7: Restart HA

Use mcp__home-assistant__ha_restart with confirm=true.

Step 8: Wait and Verify

Wait 2-3 minutes for HA to restart, then verify the deployment:

• Check that sensor.ura_presence_coordinator_presence_house_state is available
• Report the current house state and zone statuses

Step 9: Report

Print a summary:

• Version deployed
• PR URL
• Release URL
• HA status after restart