MCP Catalogs
HomeLambda-MCP-Server screenshot

Lambda-MCP-Server

by mikegc-aws·232·Score 46

AWS Lambda serverless MCP server implementation with session management and HTTP streaming support.

developer-toolsops-infraai-llm
48
Forks
3
Open issues
11 mo ago
Last commit
2d ago
Indexed

Overview

The Lambda-MCP-Server is a production-ready Python implementation that enables developers to deploy MCP servers on AWS Lambda with minimal boilerplate code. It provides a simple decorator-based approach for creating tools, built-in session state management using DynamoDB, and supports streamable HTTP for efficient communication. The server is actively maintained and provides example implementations for weather queries, S3 bucket counting, and time retrieval functions.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

you:Deploying serverless MCP tools on AWS infrastructure
you:Creating stateless AI applications with persistent session context
you:Building scalable MCP endpoints for cloud-based AI workflows
you:What authentication methods are supported?
you:How is session state managed?

When to choose this

Choose this when you need to deploy MCP tools serverlessly on AWS and want minimal boilerplate with good developer experience.

When NOT to choose this

Avoid if you're not using AWS ecosystem, as this is tightly coupled to AWS Lambda and related services.

Tools this server exposes

5 tools extracted from the README
  • get_weatherget_weather(city: str)

    Get the current weather for a city.

  • count_s3_bucketscount_s3_buckets()

    Count the number of S3 buckets in your AWS account.

  • get_timeget_time()

    Get the current UTC date and time.

  • increment_counterincrement_counter()

    Increment a session-based counter.

  • get_counterget_counter()

    Get the current counter value.

Comparable tools

mcp-server-basefastmcpmcp-aws-lambda

Installation

Install via pip:

pip install awslabs.mcp_lambda_handler

For Claude Desktop, add to your config.json:

{
  "mcpServers": {
    "lambda-mcp": {
      "command": "python",
      "args": ["-m", "awslabs.mcp_lambda_handler"],
      "env": {
        "AWS_ACCESS_KEY_ID": "YOUR_AWS_KEY",
        "AWS_SECRET_ACCESS_KEY": "YOUR_AWS_SECRET"
      }
    }
  }
}

FAQ

What authentication methods are supported?
The server supports Bearer token authentication via API Gateway authorizers. It can be extended to support AWS IAM, OAuth 2.0, or Amazon Cognito for production use.
How is session state managed?
Session state is automatically managed using a DynamoDB table with the session ID as the key. Session data persists across tool invocations within the same conversation.

Compare Lambda-MCP-Server with

GitHub →

Last updated · Auto-generated from public README + GitHub signals.