S3 MCP Server

Name: S3 MCP Server
Author: samuraikun

@samuraikun

Repository

About S3 MCP Server

MCP server to integrate AWS S3 and LLM

#aws #typescript #mcp #aws-s3 #mcp-servers #modelcontextprotocol

Basic information

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "aws-s3-mcp": {
      "command": "npx",
      "args": [
        "aws-s3-mcp"
      ]
    }
  }
}

Tools

No tools detected

We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.

Overview

What is S3 MCP Server?

An Amazon S3 Model Context Protocol (MCP) server that provides tools for LLMs like Claude to interact with S3 buckets and objects. It supports listing buckets, listing objects, and retrieving object contents.

How to use S3 MCP Server?

Install via npm (npm install -g aws-s3-mcp), configure AWS credentials and environment variables such as AWS_REGION, S3_BUCKETS, and S3_MAX_BUCKETS, then run with HTTP transport (default) or STDIO transport. For Claude Desktop, run the server with the --stdio flag and add it to the Claude Desktop configuration.

Key features of S3 MCP Server

Supports STDIO, HTTP, and Streamable HTTP transports
Tools: list-buckets, list-objects, get-object
Deployment: local Node.js, Docker CLI, Docker Compose (with MinIO)
Built-in MCP Inspector for debugging and testing
Configurable allowed buckets and region

Use cases of S3 MCP Server

Allow Claude or other LLMs to browse S3 buckets and retrieve objects
Integrate S3 access directly into Claude Desktop workflows
Test S3 interactions locally using MinIO with Docker Compose
Debug MCP server behavior with the built-in inspector tool

FAQ from S3 MCP Server

What transports does the server support?

The server supports STDIO, HTTP, and Streamable HTTP transports. HTTP is the default transport; STDIO is required for Claude Desktop integration.

How do I configure AWS credentials?

Set environment variables AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY, and AWS_REGION, or rely on the default AWS credentials file. A .env file can also be used.