EchoFlow
Provides integration with Amazon's Alexa+ voice assistant, enabling autonomous smart home control, daily briefings, routine orchestration, and knowledge base queries through MCP tools.
This server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
โก EchoFlow: Autonomous Alexa+ MCP Agent with AWS Bedrock
Submitted to: Build, Ship, Shape: Amazon Developer Hackathon 2026
Primary Track: Alexa+ Track (Streamable HTTP Model Context Protocol spec2025-11-25+)
Mini-Challenges: AWS Builder (Bedrock Runtime) & Open Source (MIT License)
Judging Bonus: Up to +10% Bonus via Comprehensive Developer Friction Log
๐ Overview
EchoFlow is a next-generation proactive voice assistant agent bridge for Alexa+. Built on the open Model Context Protocol (MCP) and powered by AWS Bedrock (Anthropic Claude 3.5 Sonnet / Amazon Nova), EchoFlow transforms Alexa+ from a passive command speaker into an autonomous smart life agent that can reason across devices, synthesize proactive daily briefings, and orchestrate complex multi-step routines in real-time.
flowchart TD
subgraph Client["Interactive Web Companion (Port 5173)"]
UI["React 18 + Tailwind Dashboard"]
Voice["Web Speech Voice Engine & Waveform Visualizer"]
Inspector["Live MCP SSE Protocol Inspector"]
Sandbox["Smart Home Virtual Device Sandbox"]
end
subgraph Server["Alexa+ MCP Server (Port 3001)"]
HTTP["Streamable HTTP Transport (/sse, /messages, /tools)"]
MCP["MCP Server Protocol Engine (@modelcontextprotocol/sdk)"]
Tools["MCP Tool Registry (Smart Home, Briefings, Routines, KB)"]
Bedrock["AWS Bedrock Reasoning Agent"]
end
Voice -->|Voice Query| UI
UI -->|POST /messages| HTTP
HTTP --> MCP
MCP --> Bedrock
Bedrock --> Tools
Tools -->|SSE Stream Events| Inspector
Inspector --> UIRelated MCP server: AlexaTasker
โจ Key Features
Official Streamable HTTP MCP Server (
spec 2025-11-25+):Implements Server-Sent Events (SSE) on
/sseand JSON-RPC tool endpoints on/messages.Exposes 4 production-grade MCP tools:
get_daily_briefing: Synthesizes contextual weather, schedule, energy metrics, and voice scripts.control_smart_device: Controls lights, eco-thermostats, smart locks, and security sensors.trigger_ops_routine: Coordinates multi-step composite routines (Morning Launch, Focus Mode, Away Lockdown).query_knowledge_base: Queries user context, contacts, and preferences.
AWS Bedrock Intelligence Layer:
Uses Anthropic Claude 3.5 Sonnet and Amazon Nova for multi-hop tool execution and natural voice generation.
Includes an intelligent zero-config local engine for instantaneous offline testing and screen recordings.
Interactive Alexa+ Web Companion & Visual Simulator:
Real-time speech recognition and text-to-speech audio synthesis.
Dynamic audio wave visualizer.
Live MCP Protocol Inspector: Real-time stream trace of every JSON-RPC request, response, and latency metric.
Smart Home Sandbox: Interactive device controls with instant visual state feedback.
1-Click Demo Scenarios: Instant execution of key demo workflows for hackathon judges.
10% Judging Bonus Documentation:
Full developer friction log and actionable platform suggestions documented in
FRICTION_LOG.md.
๐ Quick Start Guide
Prerequisites
Node.js
v18+orv20+orv22+/v26+npm
v9+
1. Installation
Clone the repository and install all dependencies:
git clone https://github.com/Ayoola-tech2024/echoflow-alexa-mcp.git
cd echoflow-alexa-mcp
npm install2. (Optional) Configure AWS Bedrock Credentials
Create a .env file in the server directory if using live AWS Bedrock:
AWS_REGION=us-east-1
AWS_ACCESS_KEY_ID=your_access_key_id
AWS_SECRET_ACCESS_KEY=your_secret_access_key
PORT=3001(Note: If omitted, EchoFlow automatically operates in instant local agentic mode with zero setup required!)
3. Run EchoFlow (Server + Web Simulator)
Start both backend MCP server and frontend simulator concurrently:
npm run devAlexa+ MCP Server:
http://localhost:3001Health Endpoint:
http://localhost:3001/healthStreamable HTTP (SSE):
http://localhost:3001/sseTools List:
http://localhost:3001/tools
Web Companion Simulator:
http://localhost:5173
๐งช Running Automated Tests
Run the full Jest test suite verifying MCP protocol handlers, health endpoints, and agent queries:
npm test๐ Repository Structure
echoflow-alexa-mcp/
โโโ LICENSE # MIT Open Source License
โโโ README.md # Project documentation & architecture
โโโ CONTRIBUTING.md # Open-source contribution guide
โโโ FRICTION_LOG.md # Developer friction log (+10% bonus)
โโโ DEMO_SCRIPT.md # 3-minute video recording script
โโโ DEVPOST_SUBMISSION.md # Devpost copy-paste submission pack
โโโ package.json # Root monorepo orchestration
โโโ tsconfig.json # Root TypeScript configuration
โโโ shared/
โ โโโ types.ts # Shared TypeScript contracts
โโโ server/ # Alexa+ MCP Server & AWS Bedrock
โ โโโ package.json
โ โโโ src/
โ โ โโโ index.ts # Express + Streamable HTTP SSE server
โ โ โโโ mcp-server.ts # MCP Protocol Server & handlers
โ โ โโโ state.ts # In-memory environment state manager
โ โ โโโ bedrock/
โ โ โ โโโ bedrock-client.ts # AWS Bedrock client & agent engine
โ โ โโโ tools/
โ โ โโโ briefing.ts # get_daily_briefing tool
โ โ โโโ smart-home.ts # control_smart_device tool
โ โ โโโ routines.ts # trigger_ops_routine tool
โ โ โโโ knowledge-base.ts # query_knowledge_base tool
โ โโโ test/
โ โโโ mcp.test.ts # Jest MCP test suite
โโโ client/ # Interactive Alexa+ Web Simulator
โโโ package.json
โโโ vite.config.ts
โโโ index.html
โโโ src/
โโโ App.tsx # Main layout
โโโ components/
โ โโโ Header.tsx
โ โโโ VoiceSimulator.tsx
โ โโโ McpInspector.tsx
โ โโโ SmartHomeSandbox.tsx
โ โโโ ScenarioPlayer.tsx
โโโ services/
โโโ api.ts # REST & SSE stream client๐ License
This project is open-source software licensed under the MIT License.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelogโฆ
This server cannot be deployed
Maintenance
Related MCP Connectors
One place to build, share, and govern the skills and tools your AI agents use at work.
Connect your AI agent to 300+ methods and 100+ live tools
Connect AI agents to 1000+ apps with managed authentication and tool-calling.
Context and control for any AI. Agent identity, scoped access, and proof across tools.
Related MCP Servers
- FlicenseNot gradedqualityDmaintenanceEnables AI agents to control Alexa-connected smart home devices, including voice announcements, music control, smart lighting, sensor monitoring, and volume management through the Alexa API.-
- AlicenseNot gradedqualityBmaintenanceEnables Alexa+ to manage your tasks and calendar through conversational commands, including adding, listing, completing tasks, and generating AI-powered daily plans with AWS Bedrock.5 npm1MIT
- AlicenseNot gradedqualityBmaintenanceEnables an Alexa-style assistant to plan and execute multi-step smart home routines, remember user preferences across sessions, and render interactive UI cards on screen via Streamable HTTP MCP tools. Uses AWS Bedrock for reasoning with a built-in offline intent fallback.MIT
- AlicenseNot gradedqualityCmaintenanceEnables Alexa+ to call a curated set of Streamable HTTP tools that bridge voice requests to a persistent personal AI agent for asynchronous tasks, memory, home control, calendar, GitHub, and messaging.MIT