Home Assistant Manager
Edit, deploy, and verify Home Assistant YAML and Lovelace dashboards over SSH/hass-cli/MCP
Test report
- Verdict
- Tested · Works
- Score
- Tested
- Jul 21, 2026
- Environment
- Claude Code 2.x (agent harness)
- Upstream re-checked
- Aug 10, 2026 · 1411a12
Fetched SKILL.md and both reference files via raw GitHub — all three (reference/automations.md, reference/dashboards.md, dashboard.png) returned HTTP 200; frontmatter parses with name+description; no hardcoded machine paths (root@homeassistant.local is explicitly documented as a placeholder) and no security smells. For OUTPUT I wrote two real automation.yaml artifacts for the same task ("porch light on when front door opens after sunset, plus deploy steps"): baseline.yaml from memory vs skill.yaml following the body. Concrete diffs: baseline emitted legacy `trigger:/condition:/action:` + `platform: state` + `service:`, omitted a stable `id:` and `mode:`, and said "restart Home Assistant"; the skill version produced modern plural `triggers:/conditions:/actions:` + `trigger: state`/`action:`, added `id: porch_light_front_door_after_sunset` and `mode: single`, and correctly deployed via reload of the automation domain (never restart for an automation edit) with a manual-trigger + log-grep + state-check verify loop — the reload-vs-restart correction is the most consequential difference.
Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works
- Installs cleanly 5/5
- Triggers reliably 5/5
- Output vs. baseline 8/10
- Docs & honesty 5/5
What Home Assistant Manager does
Guides Claude through editing Home Assistant configuration (automations, scripts, scenes, templates, MQTT, Lovelace dashboards), validating with `ha core check`, deploying via git pull or rapid scp, choosing reload vs restart correctly, and verifying changes from logs, traces, and entity state. Triggers on any Home Assistant config, automation, template, or dashboard task done over SSH, hass-cli, or an MCP server. Enforces modern 2024.10+ automation syntax, stable automation ids, and type-coerced templates.
How to install Home Assistant Manager
git clone --depth 1 https://github.com/komal-SkyNET/claude-skill-homeassistant.git /tmp/home-assistant-manager-src
mkdir -p ~/.claude/skills
cp -R /tmp/home-assistant-manager-src/skills/home-assistant-manager ~/.claude/skills/home-assistant-manager
# SKILL.md lands at ~/.claude/skills/home-assistant-manager/SKILL.md, plus reference/ (automations.md, dashboards.md) and dashboard.png.
# Runtime prerequisites (the skill drives these, does not bundle them): a reachable Home Assistant instance and ONE of:
# - SSH access to the HA host (for `ha core check|restart|logs`), or
# - hass-cli with HASS_SERVER + HASS_TOKEN exported BEFORE the session starts, or
# - an HA MCP server (official mcp_server integration or community ha-mcp).
# The repo you edit is assumed to be the HA /config dir, git-connected to the instance.
# Plugin-marketplace alternative: repo ships .claude-plugin/marketplace.json + plugin.json.
Skills live in ~/.claude/skills/ (global) or .claude/skills/
(per-project). Restart Claude Code after installing.
Commands — how to trigger Home Assistant Manager
-
/home-assistant-managerEdit, deploy, and verify Home Assistant YAML and Lovelace dashboards over SSH/hass-cli/MCP
It also activates on plain-language prompts like these:
-
Add a new automation to Home Assistant and deploy it -
Validate this Home Assistant blueprint before deploying -
Build a Lovelace dashboard for my sensors
Frequently asked questions
- Is the Home Assistant Manager skill free?
- Yes. The skill itself is free from komal-SkyNET/claude-skill-homeassistant. SkillProof publishes the install command and an independent test verdict at no cost.
- Does Home Assistant Manager work with Claude Code?
- We tested it with Claude Code 2.x (agent harness) on Jul 21, 2026. Verdict: Tested · Works. Fetched SKILL.md and both reference files via raw GitHub — all three (reference/automations.md, reference/dashboards.md, dashboard.png) returned HTTP 200; frontmatter parses with name+description; no hardcoded machine paths (root@homeassistant.local is explicitly documented as a placeholder) and no security smells. For OUTPUT I wrote two real automation.yaml artifacts for the same task ("porch light on when front door opens after sunset, plus deploy steps"): baseline.yaml from memory vs skill.yaml following the body. Concrete diffs: baseline emitted legacy `trigger:/condition:/action:` + `platform: state` + `service:`, omitted a stable `id:` and `mode:`, and said "restart Home Assistant"; the skill version produced modern plural `triggers:/conditions:/actions:` + `trigger: state`/`action:`, added `id: porch_light_front_door_after_sunset` and `mode: single`, and correctly deployed via reload of the automation domain (never restart for an automation edit) with a manual-trigger + log-grep + state-check verify loop — the reload-vs-restart correction is the most consequential difference.
- What is the Home Assistant Manager SkillProof Score?
- 9.2/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 8/10, docs & honesty 5/5.
- How do I install Home Assistant Manager?
- Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
- Can I use Home Assistant Manager with Cursor, Copilot, Gemini CLI, Codex or other AI tools?
- The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.