awslabs

mcp: Connect AI Tools to AWS with Model Context Protocol

awslabs/mcp is a suite of AWS-focused servers built on the Model Context Protocol (MCP), the open protocol created by Anthropic. It gives MCP-compatible AI clients tools and AWS context for documentation, infrastructure, data, and other cloud workflows.

ActivePythonApache-2.0AI AgentsMCPMCP Server
Stars
9.9k
Forks
1.9k
Stars, last 30 days
+141
Last commit
today
Contributors
100+

Our take

Activeregular commits and releases

Health score

63/100

How it is scored
Activity
88
Community
92
Issues
21
Pull requests
52

+141 stars in the last 30 days

  • Releases more often than 92% of the projects we track
  • Merges more pull requests than 83% of the projects we track
  • More contributors than 78% of the projects we track

This is a broad, actively maintained AWS MCP catalog, but teams planning production agent workflows should evaluate the successor toolkit and account for uneven issue closure and transport support.

Good fit if

  • You need AWS-specific tools or documentation in an MCP-compatible coding assistant or application.
  • Your use case maps to one of the many specialized servers and stdio transport is suitable.
  • You want an Apache-2.0-licensed starting point and can choose permissions and AWS credentials appropriate to each server.

Look elsewhere if

  • You are building production agent software and need the successor's stated IAM condition keys, CloudWatch and CloudTrail visibility, or evaluated skills. The README recommends the AWS Agent Toolkit for that use.
  • Your MCP client requires SSE or Streamable HTTP from these repository servers. The README says SSE was removed and the servers are designed for stdio.
  • You need confidence from a stated test requirement or consistently demonstrated issue closure. The supplied project facts report no test requirement and no issues closed in the measured period.
All health signals
Last commit2026-11-08 (today)
Commits, last 90 days100+
Releases, last 12 months92 (latest 2026.10.20261030201955, 2026-10-30)
Contributors100+ (top contributor: 21% of commits)
Issues closed, last 90 days0+
Pull requests merged, last 90 days74 (typically merged in 2 days)
Project age20 months

Checked on 2026-11-09 with the GitHub API.

Project history

Daily GitHub snapshots since Oct 7, 2026. History covers stars and forks; health signals are not tracked historically.

Selected snapshot

Stars

9,929

Forks

1,881

Change since Nov 16, 2026: +1 stars·+2 forks

Overview

awslabs/mcp is a suite of specialized servers built on the Model Context Protocol (MCP), an open protocol created by Anthropic. The servers connect MCP-compatible AI applications, including coding assistants and chat tools, with AWS documentation, services, and workflows so those applications can use relevant cloud context and tools.

The collection spans documentation, infrastructure, containers, serverless, AI services, and databases. The repository's servers support stdio transport; it also lists AWS-hosted remote servers. The README identifies the AWS Agent Toolkit as the successor for production agent use, while noting that this repository remains available and accepts contributions.

Key Features

  • Provides a catalog of specialized servers for AWS documentation, infrastructure, deployment, AI, data, and operations.
  • Connects MCP-compatible hosts and clients to server capabilities through the standardized MCP architecture.
  • Offers quick-install configurations for clients including Kiro, Cursor, and VS Code.
  • Includes servers for AWS documentation and knowledge, Infrastructure as Code, EKS, ECS, Lambda, and many AWS data services.
  • Supports stdio transport for the repository's servers; some AWS-hosted services are listed as remote servers.
  • Includes examples for configuring servers in multiple AI clients and links to server-specific setup instructions.

Use Cases

  • A cloud developer can give a coding assistant access to current AWS documentation while designing or debugging an application.
  • An infrastructure engineer can use IaC-focused tools for CDK or CloudFormation guidance and deployment troubleshooting.
  • A team working with AWS databases can connect an assistant to a relevant database server for service-specific operations.
  • An AI application developer can select a specialized server to provide AWS service context or tools to an MCP-compatible host.

What you need

Detected in the repository

  • Python (main language on GitHub)
  • Automated checks on GitHub Actions

License in plain words

Apache-2.0permissive

  • Commercial use: yes
  • Modify and redistribute: yes
  • You must keep: the license, the NOTICE file and a note of your changes
  • Share your changes: no
  • Includes an explicit patent grant from the contributors:

A summary, not legal advice: the LICENSE file is what applies.

Getting Started

For example, run the AWS Documentation MCP Server with uvx awslabs.aws-documentation-mcp-server@latest and configure it in your MCP client. See the README and the individual server guides for setup, AWS credentials, permissions, and client-specific configuration.

Alternatives

  • mcp-server-cloudflare: It connects MCP clients to Cloudflare products rather than providing AWS-focused documentation, infrastructure, and data tools.
ProjectLanguageLicenseStarsStatus
mcpPythonApache-2.09.9kActive
mcp-server-cloudflareTypeScriptApache-2.04.4kActive

Considerations

  • The repository is active, with regular commits and releases and a broad contributor base; the measured contributor share is not concentrated in one person. However, the supplied health data reports no issues closed in the measured period, so open issues may not be a reliable measure of response time.
  • Server capabilities, required AWS permissions, and setup vary by server. Some examples explicitly enable write or sensitive-data access, so review each server's configuration before granting access.
  • The repository's servers are designed for stdio. SSE support was removed, and the README describes Streamable HTTP support as future work.
  • The README presents the AWS Agent Toolkit as the successor and directs production builders toward it. It says useful projects from this repository may move there over time.
  • The repository reports CI but no test requirement in the supplied project facts; this is not evidence that every server lacks tests, but it offers no repository-wide test signal here.

Found this useful?

Share it with someone who would like mcp.

Comparisons