alexa-mcp

alexa-mcp

Enables AI assistants to audit and clean up Amazon Alexa accounts: list routines, smart-home devices, scenes, and activity history, with optional deletion of orphan devices.

Category
Visit Server

README

alexa-mcp

The cleanup-focused Alexa MCP server. An MCP server that lets an AI assistant (Claude Code, Claude Desktop, …) audit and clean up your Amazon Alexa account: list and inspect routines, list smart-home devices and delete the orphans that pile up over time, plus read scenes and activity history.

It talks to Amazon's internal Alexa cloud API via alexa-remote2.

⚠️ Disclaimer

This project is not affiliated with, endorsed by, or supported by Amazon. There is no official public API for managing Alexa routines or smart-home devices — this tool uses the same undocumented, reverse-engineered endpoints the Alexa app uses, and may break at any time when Amazon changes them. Using it may violate Amazon's Terms of Service. Use at your own risk, against your own account only. The login happens on Amazon's real sign-in page (including 2FA); your credentials go only to Amazon — the local proxy only captures the resulting session token afterwards.

Why this exists (vs. other Alexa MCP servers)

Most Alexa MCP servers focus on control (announcements/TTS, volume, running routines, lists, groups). This one is deliberately narrow and complementary — it focuses on the audit & cleanup gap that nothing else covers and that Amazon's own app can no longer do (bulk-deleting smart-home devices was removed from the Alexa web app):

Capability This server
List routines ✅
Inspect a routine's triggers & action sequence ✅
Delete a routine ✅ (write-gated; endpoint unverified — see limitations)
List smart-home devices (with source + reachability) ✅
Delete a smart-home device (orphan cleanup) ✅ (write-gated)
List scenes / smart-home entities ✅
Activity history ✅
Announcements / TTS / volume / lists ❌ (use a control-focused server or Home Assistant)

It is read-only by default. Destructive tools are only registered when you explicitly opt in.

Requirements

  • Node ≥ 20 (compiles with tsc to dist/, then runs on plain Node — no bundler).
  • An Amazon account with app-based 2FA (SMS/email OTP no longer works with Amazon's login).
  • Log in from a device/browser without the Alexa app installed (otherwise Amazon deep-links into the app instead of finishing the web login).

Quick start

git clone https://github.com/snordquist/alexa-mcp.git
cd alexa-mcp
npm install
npm run doctor      # shows config + the login URL (no network)
npm run auth        # ONE-TIME: starts a local proxy, browser login incl. 2FA
npm run smoke       # lists devices, routines, scenes (read-only)

npm run auth in detail

  1. It starts a local proxy and prints a URL (default http://127.0.0.1:3456/).
  2. Open that URL in a browser on the same machine and log in to your Amazon account (2FA).
  3. On success the token (cookie + refreshToken) is written to ./.auth/alexa.json (git-ignored) and refreshed automatically afterwards (every 4 days).
  4. A smoke test (device list) runs. Done.

Logging in from another device? proxyOwnIp must match the URL in your browser exactly:

ALEXA_MCP_PROXY_IP=192.168.x.x npm run auth   # then open http://192.168.x.x:3456/

Use as an MCP server

Read-only (recommended for auditing):

claude mcp add alexa -- node /ABSOLUTE/PATH/alexa-mcp/dist/index.js

With destructive tools enabled (delete routines / devices):

claude mcp add alexa --env ALEXA_MCP_ALLOW_WRITE=1 -- node /ABSOLUTE/PATH/alexa-mcp/dist/index.js

(Run npm run build first.)

Tools

Tool Purpose Writes
alexa_list_devices Registered Echos / apps / Fire TVs no
alexa_list_routines All routines: id, name, status, triggers no
alexa_get_routine One routine as raw JSON (incl. action sequence) no
alexa_list_scenes Smart-home entities incl. scenes no
alexa_get_activity Activity history (customer-history-records) no
alexa_list_smarthome_devices Smart-home devices with source (skill/Matter) + entity id — find orphans no
alexa_trigger_routine Execute a routine yes
alexa_delete_routine Delete a routine (endpoint unverified — see limitations) yes
alexa_delete_smarthome_device Delete a smart-home device (orphan cleanup) yes

Write tools are only registered when ALEXA_MCP_ALLOW_WRITE=1, and each destructive call also requires an explicit confirm: true argument.

Configuration (environment variables)

Variable Default Purpose
ALEXA_MCP_AMAZON_PAGE amazon.de Login region (e.g. amazon.com, amazon.co.uk)
ALEXA_MCP_ACCEPT_LANGUAGE de-DE Accept-Language header
ALEXA_MCP_PROXY_IP 127.0.0.1 Must match the URL you open during login
ALEXA_MCP_PROXY_PORT 3456 Proxy port for the login
ALEXA_MCP_PROXY_BIND 0.0.0.0 Proxy bind address
ALEXA_MCP_REFRESH_MS 345600000 (4 d) Token refresh interval
ALEXA_MCP_ALLOW_WRITE – 1 enables the destructive tools
ALEXA_MCP_AUTH_PATH .auth/alexa.json Token file location

Security & privacy

  • The session token is stored only in ./.auth/alexa.json (mode 0600, git-ignored). It is never logged or committed. Treat that file like a password.
  • The login proxy binds to 0.0.0.0 by default so you can log in from another device; set ALEXA_MCP_PROXY_BIND=127.0.0.1 to keep it local-only.
  • To fully sign out, delete ./.auth/alexa.json (and remove the registered "device" from Amazon → Manage Your Content and Devices).

Known limitations

  • Unofficial API — can break whenever Amazon changes it.
  • Routine deletion endpoint is unverified. alexa_delete_routine attempts DELETE /api/behaviors/v2/automations/{id}, which is not documented or confirmed. Verify the real request via browser DevTools before relying on it.
  • Device deletion may not be permanent if the underlying source still advertises the device: a skill that is still linked (and whose backend responds) or a Matter bridge will re-add it on the next discovery. Orphans whose backend is gone (unlinked skill, removed integration) stay deleted.
  • Matter duplicate devices (the same device appearing once per Alexa Echo/hub that acts as a Matter controller) are an Alexa-side artifact, not a bridge misconfiguration. Deleting the redundant copies works, but new ones can reappear if the Matter bridge is re-commissioned.

See ARCHITECTURE.md for the auth flow, endpoint reference, and implementation notes.

License

MIT © Sascha Nordquist

Recommended Servers

playwright-mcp

playwright-mcp

A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

Official
Featured
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.

Official
Featured
Local
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

graphlit-mcp-server

The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.

Official
Featured
TypeScript
Kagi MCP Server

Kagi MCP Server

An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

Official
Featured
Python
E2B

E2B

Using MCP to run code via e2b.

Official
Featured
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Exa Search

Exa Search

A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured