Home Assistant Vibecode

Enables MCP-enabled editors to analyze, design, and deploy Home Assistant automations, dashboards, and scripts via the HA Vibecode Agent.

typescript

GitHub Stars

typescript

Language

2 months ago

First Indexed

3 weeks ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "coolver-home-assistant-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@coolver/home-assistant-mcp@latest"
      ],
      "env": {
        "HA_AGENT_KEY": "your_api_key_here",
        "HA_AGENT_URL": "http://<home-assistant-host>:8099"
      }
    }
  }
}

You install and run an MCP server that talks to the Home Assistant Vibecode Agent inside your Home Assistant instance. This setup lets your AI-enabled editor connect to HA, inspect your configuration, and automatically design, deploy, and manage automations, dashboards, and scripts in a safe, versioned workflow. You control how much you delegate to AI while keeping full access to your Home Assistant environment.

How to use

Connect your MCP-enabled editor (Cursor, VS Code, or any MCP-enabled IDE) to the Home Assistant Vibecode Agent running on your Home Assistant instance. Once connected, you describe your goal in natural language. The agent analyzes your Home Assistant configuration, entities, and devices, and then the MCP server coordinates with the editor to generate intelligent automations, scripts, and dashboards. You can also use the agent as a DevOps helper to upload changes, run tests, and analyze logs.

How to install

Prerequisites and steps to set up the MCP server and agent so you can start building with AI.

Ensure Node.js is installed on the computer where your AI editor runs.
Install the HA Vibecode Agent on Home Assistant (as an add-on) and start it.
In your AI editor, configure the MCP connection to the agent using the provided configuration snippet.
Test the connection by sending a sample request from your editor to verify access to climate entities and agent status.
Begin describing your automation goals in natural language and let the AI generate and deploy the solutions onboard.

Configuration and security

Environment variables and how to configure the MCP client to reach the HA Vibecode Agent.

Environment variables shown in the example below are required for authentication and endpoint configuration.

{
  "mcpServers": {
    "home-assistant": {
      "command": "npx",
      "args": ["-y", "@coolver/home-assistant-mcp@latest"],
      "env": {
        "HA_AGENT_URL": "http://<home-assistant-host>:8099",
        "HA_AGENT_KEY": "your_api_key_here"
      }
    }
  }
}

Troubleshooting

If you encounter issues, verify that the HA Vibecode Agent is running, ensure the AI editor is fully restarted, and confirm the MCP configuration was entered correctly.

Common checks include confirming the agent health endpoint is reachable and ensuring the correct agent URL and key are used in the MCP configuration.

Security and governance

All MCP interactions route through the HA Vibecode Agent for a controlled, authenticated surface to Home Assistant. Agent keys are stored only in your local IDE configuration and should never be committed to version control.

The system provides human-readable commit messages and a full activity log so you can audit what changes the AI made and when.

Development and ecosystem

The MCP setup is designed to work with the Home Assistant Vibecode Agent and various MCP-enabled IDEs, allowing you to install, configure, and extend with community integrations and dashboards through the agent-driven workflow.

Built by

VeilStrat

AI signals for GTM teams