---
name: ha-apply-change
description: Validate and apply a generated HA automation or script YAML with safety checks and optional reload. Use after building or modifying an automation.
allowed-tools:
  - Bash
  - Read
  - Write
---

# Apply HA Change

## Steps

1. **Pre-check**: Run `${CLAUDE_PLUGIN_ROOT}/bin/ha-agent-lab ha policy-check <artifact_path>` to verify safety.
   - If blocked: stop and explain why. Create a proposal via `/claude-code-hermit:proposal-create`.

2. **Validate and apply**: Run `${CLAUDE_PLUGIN_ROOT}/bin/ha-agent-lab ha validate-apply <artifact_path> --reload automation` (or `script`).
   - This runs HA config check, **pushes the config to HA via REST**, then reloads the domain.
   - The artifact must include `id:` at the top level — if missing, the CLI derives an ID from the alias or filename and warns in the output. A derived ID drifts if the alias is renamed, creating a duplicate.

3. **Confirm with operator**: Always ask before executing the apply. Show:
   - The artifact being applied
   - Policy check result
   - What domain will be reloaded

4. **Post-apply**: Check the JSON output for `creation_ok` and read `.claude-code-hermit/raw/audit-ha-apply-latest.md`.
   - `creation_ok: true` — config was pushed and verified via REST. Reload picks it up immediately.
   - `creation_ok: false` + message contains "YAML mode" — HA is in YAML config mode (403). Tell the operator to place the generated YAML in their HA config directory and reload manually.
   - `creation_ok: false` + other message — push failed with a validation error from HA. Show the error message and suggest fixing the YAML.
   - If overall successful: update your auto memory if this automation introduces a new pattern.

## Safety

- The apply path only reloads `automation` and `script` domains.
- Sensitive entities are blocked at the policy-check step.
- The operator must confirm before any reload happens.