Lambda MCP Server Demo (Streamable HTTP)

@mikegc-aws

Basic information

About Lambda MCP Server Demo

Creates a simple MCP tool server with "streaming" HTTP.

#aws #lambda #aws-lambda #mcp#lambda-functions#amazon-web-services#model-context-protocol #mcp-server #model-context-protocol-servers

Lambda MCP Server Demo config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "Lambda-MCP-Server": {
      "command": "uv",
      "args": [
        "sync"
      ]
    }
  }
}

Lambda MCP Server Demo tools

Tool list not available yet

This listing does not currently have a tool list to display. This does not mean the server has no tools; check its documentation for supported capabilities.

We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.

Overview

What is Lambda MCP Server Demo (Streamable HTTP)?

Lambda MCP Server Demo (Streamable HTTP) is a serverless MCP (Model Context Protocol) server running on AWS Lambda. It demonstrates how to build stateless MCP tools with minimal boilerplate and an excellent developer experience, using Streamable HTTP transport instead of SSE. It is aimed at developers who want to deploy cloud-hosted MCP tools quickly.

How to use Lambda MCP Server Demo (Streamable HTTP)?

Install the package from PyPI (pip install awslabs.mcp_lambda_handler) and define tools using the mcp_server.tool() decorator. Deploy the server with AWS SAM. The client must support Streamable HTTP (e.g., MCP Inspector or the included Strands Agents client).

Key features of Lambda MCP Server Demo (Streamable HTTP)

Serverless MCP server on AWS Lambda
Uses Streamable HTTP transport (not SSE)
Minimal boilerplate with decorator-based tool creation
Built-in session state management with DynamoDB
Bearer token authentication via API Gateway custom authorizer
Automatic type validation, request parsing, and error handling

Use cases of Lambda MCP Server Demo (Streamable HTTP)

Deploying MCP tools without managing servers
Maintaining session context across tool invocations
Building cloud-hosted AI agent tool integrations
Demonstrating MCP with Streamable HTTP on AWS

Frequently asked questions

What transport does this server require?

It requires Streamable HTTP (not SSE). Clients that support Streamable HTTP include the MCP Inspector and the Strands Agents client included in this repo.

What dependencies are needed to run the server?

You need an AWS account, AWS SAM CLI, and the package `awslabs.mcp_lambda_handler`. The client also requires access to Amazon Bedrock with Amazon Nova Pro enabled.

Where does session data live?

Session data is stored in a DynamoDB table against a sessionId key. The table is managed automatically.

How is authentication handled?

Authentication uses a Bearer token passed in the `Authorization` header, validated by a custom API Gateway Lambda authorizer. The token is configured during deployment.

What are the known limitations?

This is a proof-of-concept; for production use, consider implementing more robust authentication (e.g., IAM roles, OAuth 2.0, Amazon Cognito) and following AWS security best practices.