awslabs/run-model-context-protocol-servers-with-aws-lambda

Run existing Model Context Protocol (MCP) stdio-based servers in AWS Lambda functions

What it solves

Most Model Context Protocol (MCP) servers are designed to run locally as child processes communicating via standard input/output (stdio). This project provides a way to wrap these stdio-based servers into AWS Lambda functions, allowing them to be hosted in the cloud and accessed via HTTPS instead of being tied to a single local machine.

How it works

The library manages the lifecycle of the MCP server within a Lambda function. For every invocation, the Lambda function starts the MCP server as a child process, initializes it, forwards the incoming request, returns the response, and then shuts down the process. It supports four connection methods:

  1. Amazon API Gateway: Uses the MCP Streamable HTTP transport with OAuth authentication.
  2. Amazon Bedrock AgentCore Gateway: Uses the MCP Streamable HTTP transport with OAuth authentication.
  3. Lambda Function URLs: Uses a custom Streamable HTTP transport authenticated with AWS IAM (SigV4).
  4. Lambda Invoke API: Uses a custom transport calling the Lambda API directly via AWS IAM.

Who it’s for

Developers who want to move their MCP servers from local environments to a distributed cloud architecture while maintaining compatibility with MCP clients (like Cursor, Cline, or Claude Desktop).

Highlights

  • Multi-language support: Provides implementation examples for both Python and TypeScript.
  • Cloud-native hosting: Enables MCP servers to run as serverless functions.
  • Flexible authentication: Supports OAuth (via API Gateway/Bedrock) and AWS IAM.
  • Detailed configuration: Includes guidance on packaging servers, managing secrets via AWS Secrets Manager, and passing AWS credentials to child processes.

Related

  • Project
    sam70361/aora-bot

    A zero-dependency SVG and JavaScript expression engine that provides 32 emotional states for AI assistants, chatbots, and desktop pets.

  • Project
    currents-dev/playwright-best-practices-skill

    A specialized AI skill that provides guidance and best practices for writing and maintaining Playwright tests in TypeScript across various testing modalities.

  • Project
    dorjeduck/llm.mojo

    A port of Andrej Karpathy's llm.c to the Mojo language, demonstrating that Mojo can achieve C-like performance for low-level LLM training.

  • Project
    flagos-ai/FlagPerf

    An integrated AI hardware evaluation engine that benchmarks the actual performance of AI chips across various models, frameworks, and software stacks to ensure fair and reproducible results.