Skip to content
MCP server

mcp

Collection of open source MCP servers giving AI assistants access to AWS services and docs

Actively maintainedStatus and history

The repository was last updated yesterday and is published under Apache-2.0.

Pricing
Open source
Platforms
CLI, self-hosted, vscode
GitHub stars
9.9K
Last commit
1 day ago
Language
Python
License
Apache-2.0

Similar MCP servers

About mcp

mcp is an open-source MCP server for cloud and infrastructure. It is available on CLI, self-hosted and vscode.

The README describes this as a suite of separately installable servers rather than a single program, each wrapping one part of AWS so an AI client can call it as a set of tools. Coverage runs from AWS documentation lookup and infrastructure-as-code helpers to database, container, serverless, analytics and cost-estimation servers.

Servers are launched locally over stdio, typically through uvx, and the repository notes they are designed for stdio only, with SSE support removed in recent major versions. Several servers accept flags that gate write operations and sensitive data access, and most read AWS credentials from a configured profile.

The README also points readers to AWS's newer Agent Toolkit for AWS as the successor for production work, while this repository continues to accept contributions. Teams already on AWS who use MCP-capable coding assistants are the natural audience.

What it does

Modular server suite
Separate MCP servers cover documentation, infrastructure, databases, containers, serverless and more, so you install only what you need.
Documentation access
Documentation servers let an assistant look up current AWS docs, API references and What's New posts.
Infrastructure as code helpers
Servers for CDK, CloudFormation and serverless workflows help assistants draft and troubleshoot deployments.
Local stdio transport
Servers run locally over standard input and output, launched through tools such as uvx.
Write-access flags
Some servers offer flags that gate write operations and access to sensitive data.

Good for

  • Look up current AWS API details while writing code in an AI coding assistant
  • Generate CDK or Terraform infrastructure code that follows AWS guidance
  • Query DynamoDB, Aurora or Neptune databases from a chat interface
  • Estimate AWS costs before deploying an infrastructure design
  • Inspect CloudWatch metrics and logs while troubleshooting an incident

Pricing

Open source

Apache-2.0 open source; servers run locally and use your own AWS account.

Prices change. Check awslabs.github.io before you buy.

mcp FAQ

Does the repository support remote HTTP transport?

No. The README says the MCP servers in this repository are designed to support stdio only, and SSE support was removed in recent major versions.

Is the project open source?

The repository is licensed under Apache-2.0, according to the listing details.

Which MCP clients are mentioned?

The README names Kiro, Cline, Cursor, Windsurf, Claude Desktop, Claude Code and VS Code as supported clients or examples.

Is there a newer option the README recommends?

The README recommends AWS's Agent Toolkit for AWS for production software and says the most useful projects here will move into it over time.

Is mcp free?

Yes. mcp is open source, so you can use and self-host it for free.

What are the best alternatives to mcp?

The closest alternatives to mcp are VPS MCP & Digital Secretary, Servonaut, Civo MCP by usefulapi, Linux MCP daemon (mcpd) and Yaw Labs AWS MCP.

Where this comes from

  • Description: written from the project's README and the project's website.

How AIPoweredBot checks servers