Overview

The MCP S3 Server enables AI models to interact directly with Amazon S3 object storage through the Model Context Protocol (MCP). It provides a standardized interface for AI to perform storage operations while maintaining security and scalability.

Official Sample:

Developed by AWS Samples

Key Features

📦

Object Storage Operations

Upload, download, and delete objects with ease

🗄️

Bucket Management

List and manage S3 buckets

🏷️

Metadata Handling

Manage object metadata for better organization

🛡️

Security & Scalability

Leverage S3's robust security and scalable infrastructure

Available Tools

Quick Reference

Tool	Purpose	Category
`ListBuckets`	List all S3 buckets	Discovery
`ListObjectsV2`	List objects within a bucket	Discovery
`GetObject`	Retrieve an object from S3	Read
`PutObject`	Upload an object to S3	Write
`DeleteObject`	Delete an object from S3	Write

Detailed Usage

ListBuckets▶

Returns a list of all buckets owned by the authenticated sender of the request.

use_mcp_tool({
  server_name: "s3",
  tool_name: "ListBuckets",
  arguments: {}
});

ListObjectsV2▶

Returns some or all (up to 1,000) of the objects in a bucket with each request.

use_mcp_tool({
  server_name: "s3",
  tool_name: "ListObjectsV2",
  arguments: {
    Bucket: "your-bucket-name",
    Prefix: "optional-prefix/"
  }
});

GetObject▶

Retrieves an object from Amazon S3. Specify the full key name for the object.

use_mcp_tool({
  server_name: "s3",
  tool_name: "GetObject",
  arguments: {
    Bucket: "your-bucket-name",
    Key: "path/to/your/object.txt"
  }
});

PutObject▶

Uploads an object to a specified S3 bucket.

use_mcp_tool({
  server_name: "s3",
  tool_name: "PutObject",
  arguments: {
    Bucket: "your-bucket-name",
    Key: "path/to/new/object.txt",
    Body: "Content of the object"
  }
});

DeleteObject▶

Deletes an object from a specified S3 bucket.

use_mcp_tool({
  server_name: "s3",
  tool_name: "DeleteObject",
  arguments: {
    Bucket: "your-bucket-name",
    Key: "path/to/object/to/delete.txt"
  }
});

Installation

{
  "mcpServers": {
    "s3": {
      "command": "npx",
      "args": [
        "-y",
        "aws-s3-mcp",
        "--stdio"
      ],
      "env": {
        "AWS_REGION": "your-aws-region",
        "AWS_ACCESS_KEY_ID": "your-access-key-id",
        "AWS_SECRET_ACCESS_KEY": "your-secret-access-key",
        "S3_BUCKETS": "bucket1,bucket2"
      }
    }
  }
}

Configuration:

Replace your-aws-region, your-access-key-id, your-secret-access-key, and bucket1,bucket2 with your actual AWS credentials and desired S3 buckets.

Common Use Cases

1. Storing and Retrieving User-Generated Content

Store images, videos, and documents uploaded by users.

// Upload a user profile picture
use_mcp_tool({
  server_name: "s3",
  tool_name: "PutObject",
  arguments: {
    Bucket: "user-content-bucket",
    Key: "users/user123/profile.jpg",
    Body: "base64-encoded-image-data"
  }
});

// Retrieve a user document
use_mcp_tool({
  server_name: "s3",
  tool_name: "GetObject",
  arguments: {
    Bucket: "user-content-bucket",
    Key: "users/user123/document.pdf"
  }
});

2. Data Archiving and Backup

Automate backups of critical application data to S3.

// Upload a database backup file
use_mcp_tool({
  server_name: "s3",
  tool_name: "PutObject",
  arguments: {
    Bucket: "backup-archive-bucket",
    Key: "database/daily-backup-2023-10-27.sql",
    Body: "database-dump-content"
  }
});

3. Hosting Static Websites

Serve static website assets directly from S3.

// Upload an HTML file for a static website
use_mcp_tool({
  server_name: "s3",
  tool_name: "PutObject",
  arguments: {
    Bucket: "my-static-website",
    Key: "index.html",
    Body: "<html><body><h1>Hello S3!</h1></body></html>",
    ContentType: "text/html"
  }
});

Model Context Protocol (MCP): Open Standard for AI Integration

The Model Context Protocol (MCP) is an open standard enabling AI systems to connect with diverse data sources, tools, and services, eliminating custom integrations for seamless interaction.

Setting up MCP Servers in Visual Studio Code

Model Context Protocol (MCP) enables AI models to interact with external tools and services through a unified interface. This guide will walk you through setting up MCP servers in Visual Studio Code to enhance your GitHub Copilot experience.

Redis MCP Server

Redis MCP servers enable AI models to interact with Redis databases, providing capabilities for key-value operations, caching, pub/sub messaging, and high-performance data structures.

S3 MCP Server