chore: remove deploy_updates.sh, use local HA for dogfooding

Remove SSH-based deployment script. Dogfooding now uses the local HA
instance via VS Code Dev Container or docker-compose.test.yml.

Updated: CLAUDE.md release flow, CONTRIBUTING.md, TESTING_ON_HA.md,
scripts/README.md, Makefile.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

Author: kwschulz

CLAUDE.md

@@ -144,9 +144,10 @@ ruff check .                    # Full project lint
 pytest                          # Full test suite
 ```
 
-### 2. Dogfood on Local HA (if applicable)
-- Deploy to local Home Assistant via SSH or HACS beta
-- Verify core functionality works on real hardware
+### 2. Dogfood on Local HA
+- Ask the user to launch HA via VS Code task ("HA: Start (Fresh)") or `make docker-start`
+- User verifies: integration loads, sensors created, no blocking I/O warnings in logs
+- **Never deploy automatically** — no SSH, no SCP, no remote scripts
 - Fix any issues found, commit to release branch
 
 ### 3. Merge to Main

CONTRIBUTING.md

@@ -225,18 +225,17 @@ Make sure all tests pass before submitting your changes.
 pytest tests/ -v
 ```
 
-### 5. Deploy for Manual Testing (Optional)
+### 5. Test on Local HA (Optional)
 
-You can deploy your changes to a real Home Assistant instance for manual testing.
+You can test your changes on a local Home Assistant instance via Docker:
 
 ```bash
-# Interactive mode - guides you through deployment options
-./scripts/deploy_updates.sh
+# Start HA with integration bind-mounted
+make docker-start
 
-# Or specify directly:
-./scripts/deploy_updates.sh --local ~/homeassistant/config
-./scripts/deploy_updates.sh --ssh root@192.168.1.100
-./scripts/deploy_updates.sh --docker homeassistant --restart
+# Open http://localhost:8123 and add the integration
+# After code changes, restart to pick them up:
+make docker-restart
 ```
 
 See [Testing on HA](./docs/setup/TESTING_ON_HA.md) for detailed instructions and troubleshooting.

Makefile

@@ -122,10 +122,6 @@ validate-host:
 validate-ci:
 	@./scripts/ci-check.sh
 
-# Deploy to Home Assistant
-deploy:
-	@bash scripts/maintenance/deploy_updates.sh
-
 # Sync version numbers
 sync-version:
 	@python3 scripts/maintenance/update_versions.py

docs/setup/TESTING_ON_HA.md

@@ -1,37 +1,43 @@
 # Testing on Home Assistant
 
-This guide explains how to deploy a development branch of Cable Modem Monitor to a real Home Assistant instance for testing.
+This guide explains how to test a development branch of Cable Modem Monitor on Home Assistant.
 
 > **For local development** (running tests, working on code): See [Getting Started](./GETTING_STARTED.md)
 >
 > **For stable installation** (released versions): See [README.md Installation](../README.md#installation)
 
 ---
 
-## Quick Start
+## Quick Start (Docker — Recommended)
 
-### Using the Deploy Script (Recommended)
+The VS Code Dev Container or `docker-compose.test.yml` runs a real HA instance with your code bind-mounted:
 
 ```bash
-# Clone the repo (or switch to the branch you want to test)
-git clone https://github.com/solentlabs/cable_modem_monitor.git
-cd cable_modem_monitor
-git checkout feature/your-branch-name
+# From VS Code Dev Container:
+# Ctrl+Shift+P → Tasks: Run Task → "HA: Start (Fresh)"
 
-# Run the deployment script
-./scripts/maintenance/deploy_updates.sh
+# Or from the command line:
+make docker-start
 ```
 
-The script supports multiple deployment methods:
-- **Local path**: Direct copy to a local HA config directory
-- **SSH**: Remote deployment via SSH
-- **Docker**: Copy into a running Docker container
+Then open http://localhost:8123, complete onboarding, and add the integration.
+
+Your `custom_components/cable_modem_monitor/` is mounted directly — changes are reflected after restarting HA:
+
+```bash
+make docker-restart
+```
+
+View logs:
+```bash
+make docker-logs
+```
 
 ---
 
 ## Manual Deployment
 
-If the script doesn't work for your setup, follow these manual steps.
+For testing on a remote or standalone HA instance.
 
 ### Step 1: Get the Code
 
@@ -150,15 +156,15 @@ To test a specific pull request:
 git fetch origin pull/44/head:pr-44
 git checkout pr-44
 
-# Deploy
-./scripts/maintenance/deploy_updates.sh
+# Use docker-compose or copy manually (see above)
+make docker-start
 ```
 
 Or use GitHub CLI:
 
 ```bash
 gh pr checkout 44
-./scripts/maintenance/deploy_updates.sh
+make docker-start
 ```
 
 ---

scripts/README.md

@@ -44,7 +44,6 @@ Scripts for maintaining the integration in production:
 | Script | Purpose | Exit Codes | Environment Variables |
 |--------|---------|------------|----------------------|
 | `cleanup_entities.py` | Cleans up Home Assistant entities | 0: Success, 1: Error | None |
-| `deploy_updates.sh` | Deploy to Home Assistant via SSH | 0: Success, 1: Error | SSH configured for `homeassistant` host |
 | `update_versions.py` | Sync version from const.py to manifest.json | 0: Success, 1: Error | None |
 
 ## Usage
@@ -122,14 +121,6 @@ code .
 
 ### Maintenance Operations
 
-**Deploy to Home Assistant:**
-```bash
-# SSH must be configured for 'homeassistant' host
-bash scripts/maintenance/deploy_updates.sh
-# OR use Make
-make deploy
-```
-
 **Version Management:**
 ```bash
 # Sync version from const.py to manifest.json and hacs.json
@@ -171,19 +162,12 @@ pre-commit run --all-files
 make check
 ```
 
-## SSH Configuration for Deployment
+## Testing on Home Assistant
 
-To use `deploy_updates.sh`, configure SSH access to your Home Assistant server:
-
-```bash
-# Add to ~/.ssh/config
-Host homeassistant
-    HostName 192.168.1.100  # Your HA server IP
-    User your-username
-    IdentityFile ~/.ssh/id_rsa
-```
+Use the Docker-based local HA instance for testing. See [TESTING_ON_HA.md](../docs/setup/TESTING_ON_HA.md) for details.
 
-Test connection:
 ```bash
-ssh homeassistant "echo Connection successful"
+make docker-start    # Start HA with integration mounted
+make docker-logs     # View logs
+make docker-restart  # Restart after code changes
 ```