AWS MCP Servers

Overview

AWS MCP Servers bring the full power of Amazon Web Services to AI agents. Maintained by AWS Labs, these servers let agents manage infrastructure through natural language — from S3 operations to Lambda deployments to CloudFormation management.

How It Works

1. Install — Choose which AWS service servers you need
2. Configure IAM — Set up roles with appropriate permissions
3. Connect — Add to your MCP client configuration
4. Manage — “Deploy this Lambda” or “Show me S3 usage”

Use Cases

• Infrastructure management — Create and monitor AWS resources
• Deployment automation — Deploy via CloudFormation or CDK
• Cost analysis — Query cost and usage data
• Troubleshooting — Inspect logs, metrics, configurations
• Security auditing — Review IAM policies and access patterns

Getting Started

{
  "mcpServers": {
    "aws-s3": {
      "command": "uvx",
      "args": ["awslabs.s3-mcp-server@latest"],
      "env": { "AWS_PROFILE": "my-profile", "AWS_REGION": "us-east-1" }
    }
  }
}

Example

User: "List S3 buckets and show which are publicly accessible"

Agent → AWS S3 MCP:
  - list_buckets() → get_bucket_policy() for each

Agent: "12 buckets. ⚠️ 'backup-data' has public list enabled — 
this looks unintentional. Recommend enabling Block Public Access."

Alternatives

• Composio — AWS via managed OAuth
• Terraform MCP — Infrastructure-as-code
• Google Cloud MCP — Equivalent for GCP