Alexa

You can connect your Alexa-connected devices to AI agents through this MCP server, giving you automated control over voice announcements, music, lighting, sensors, and more. It acts as a bridge between Alexa and the MCP ecosystem, letting you orchestrate your smart home with AI agents while keeping everything centralized and self-documented for easy use.

How to use

Use an MCP client to connect to the Alexa MCP server and start issuing capabilities like announcements, music status queries, lighting control, sensor reading, and volume management. The server exposes a straightforward API surface designed for AI agents to request actions and receive status updates in real time. You can orchestrate routines across devices, trigger voice messages, adjust lighting colors and brightness, and monitor sensors from your preferred MCP client.

How to install

pnpm install

pnpm run get-cookies

pnpm run dev:node

The server will be available at http://localhost:3001

## Configuration and usage notes

Configure your MCP client to point to the server using the following connection options. You can run the server locally in stdio mode or access it as an HTTP endpoint during development.

{ "mcpServers": { "alexa": { "type": "http", "name": "alexa", "url": "http://localhost:3001", "args": [] } } }

## Security and environment variables

During setup you will work with tokens from the Alexa login flow. The environment holds tokens and base API references used by the server.

Environment variables you will encounter include tokens and a base API reference. For local development, you typically set API\_BASE to point at the local MCP API, and you store the Alexa login tokens securely.

## Troubleshooting

If the server doesn’t respond at http://localhost:3001, verify that the dev server is running and that your environment variables are loaded correctly. Check that the cookie collection script completed and that the generated .env includes UBID\_MAIN and AT\_MAIN tokens. Ensure the MCP client URL matches the API\_BASE used by the server. When issues arise, restart the server and re-authenticate if needed.

## Advanced usage

You can customize how the AI agents interact with Alexa features via the API handlers. These handlers are designed to be self-documenting so you can easily discover which actions are available (for example, announcing messages, querying current music status, or adjusting lighting). Use the MCP client to discover and invoke these capabilities in your automation flows.

## Architecture

This server is built with a fast web framework and TypeScript. It integrates MCP SDK to enable seamless cooperation with AI agents and runs in environments such as Cloudflare Workers or standard Node.js runtimes via Docker.

## Notes

If you need to enable remote access or deploy to a cloud environment, you can adapt the HTTP endpoint configuration or run the local stdio version in your deployment workflow. Remember to maintain secure handling of authentication tokens and API base URLs.

## Available tools

### VoiceAnnouncements

Send customized voice messages to Alexa devices with optional night-time suppression.

### MusicControl

Query real-time current track status and playback on Echo devices.

### SmartLighting

Control power, brightness, and colors for discovered smart lights.

### SensorIntegration

Access temperature, light, and motion sensor data from Alexa devices.

### VolumeManagement

Set and adjust volume levels across all connected devices.

### SelfDocumenting

API handlers are descriptive and easy for AI agents to discover and use.