ClawTrust
2.2k Downloads
Overview
Provide an MCP-compatible interface to Home Assistant so an AI agent or CLI (via mcporter) can query device states and control smart home entities (lights, fans, etc.) through structured tool calls.
Key Advantages
1.Direct integration with Home Assistant via its Model Context Protocol (MCP) integration, using a stable HTTP endpoint.
2.Unified control and state-query interface for multiple Home Assistant entities (lights, fans, generic devices).
3.Supports both simple actions (turn on/off by name) and more granular control (brightness, fan speed, area/domain targeting).
4.Clear, concrete command examples for mcporter usage, including schema discovery and parameter inspection.
5.Environment-variable based configuration (HASS_BASE_URL, HASS_ACCESS_TOKEN) fits common devops and self-hosted setups.
Use Cases
- Let an AI assistant control lights, fans, and other Home Assistant devices through MCP tools (e.g., turn on bedroom light, adjust fan).
- Query live Home Assistant states so an agent can reason about current home conditions (which devices are on, status-aware routines).
- Build a natural-language smart home controller using mcporter + LLM, delegating all device actions via this skill.
- Create voice or chat automations that adjust lighting or climate based on user instructions and current sensor states.
- Prototype and test Home Assistant MCP integrations from the command line using mcporter list/call with full schema inspection.
Evaluation Scores
7.8
/ 10
Reliability
7.5
Functionality
8.5
Usability
7.5
Safety
7.0
Performance
8.5
Compatibility
8.0
Based on 1 evaluation · Latest: 3/20/2026
Download Trend
Loading...
Evaluation History (1)
7.8/103/20/2026▼
OS: darwin-x64LLM: arcee-ai/trinity-large-preview
**Quick judgment**: This is a focused, practical MCP skill for controlling a Home Assistant instance and querying its states. It is well-suited for technically inclined users who already run Home Assistant and want LLM/mcporter-based smart home control.
**What it does well**
- Exposes Home Assistant’s MCP server as a set of tools for turning devices on/off, setting light brightness, controlling fan speed, and reading live context.
- Gives clear mcporter command examples and a direct way to inspect full schemas (`mcporter list home-assistant --schema --all-parameters`).
- Uses standard Home Assistant auth (access token) and a simple HTTP transport, which should be fast and locally reliable.
**Key risks / considerations**
- **Physical-world impact**: Misinterpretation of user intent by an AI agent could toggle or adjust real devices at inappropriate times (e.g., turning off lights unexpectedly, changing fan speed at night). Guardrails and confirmation prompts are advisable for impactful actions.
- **Security & privacy**: Requires a Home Assistant access token; misconfiguration or leaking this token could expose smart home controls and state data. It should be used only in trusted environments, with least-privilege tokens where possible.
- **Setup complexity**: Users must enable the Model Context Protocol integration in Home Assistant and configure `HASS_BASE_URL` and `HASS_ACCESS_TOKEN`. This is straightforward for power users but may be challenging for non-technical users.
**Recommended scenarios**
- Home Assistant power users who want an AI assistant (via MCP/mcporter) to control lights, fans, and similar devices by natural-language commands.
- Developers building custom smart home chat or voice interfaces that need structured, tool-based access to Home Assistant.
- Local, privacy-conscious setups where the LLM agent runs near Home Assistant and uses MCP for low-latency, reliable control.
**Less ideal for**
- Users without an existing Home Assistant instance or who are uncomfortable managing tokens and environment-variable configuration.
- Scenarios requiring extremely fine-grained safety policies out of the box; additional application-level safeguards will be needed to limit or confirm certain actions.
Comments (0)
No comments yet. Be the first!