AWS S3 Access MCP Server

Cloud PlatformsPython

Access and retrieve documents from Amazon S3 buckets

Available Tools

ListBuckets

Returns a list of all S3 buckets owned by the authenticated user

ListObjectsV2

Returns up to 1,000 objects in a specified S3 bucket

GetObject

Retrieves an object (document) from Amazon S3 by specifying the full key name

AWS S3 Access is a Model Context Protocol server that enables AI assistants to interact with Amazon S3 storage. It allows for listing buckets, browsing objects, and retrieving documents (currently supporting PDF files) directly from S3 buckets, making it easy to incorporate cloud-stored data into AI conversations. With this integration, AI assistants can access up to 1,000 objects from S3 buckets, providing a seamless way to work with documents stored in AWS cloud infrastructure. The server handles authentication through standard AWS credentials, making it a secure and efficient solution for organizations that rely on S3 for document storage.

Installation

To use the AWS S3 Access MCP server, you'll need to install it and configure your AWS credentials.

Prerequisites

Python 3.x installed on your system
AWS account with appropriate S3 permissions
AWS credentials configured on your machine

Installing the Server

You can install the server using pip:

pip install s3-mcp-server

Alternatively, you can install it using uv:

uv pip install s3-mcp-server

AWS Credentials Setup

Before using the server, you need to configure your AWS credentials:

Obtain your AWS access key ID, secret access key, and preferred region from the AWS Management Console
Configure the credentials using the AWS CLI or by creating credentials files:
- On Linux/macOS: ~/.aws/credentials and ~/.aws/config
- On Windows: %USERPROFILE%\.aws\credentials and %USERPROFILE%\.aws\config

Ensure your credentials have appropriate READ/WRITE permissions for S3 buckets.

Configuration in Claude Desktop

To use this MCP server with Claude Desktop, you need to update the Claude Desktop configuration file:

On MacOS: Edit ~/Library/Application\ Support/Claude/claude_desktop_config.json

On Windows: Edit %APPDATA%/Claude/claude_desktop_config.json

Add the following configuration:

{
  "mcpServers": {
    "s3-mcp-server": {
      "command": "uvx",
      "args": [
        "s3-mcp-server"
      ]
    }
  }
}

For development or custom installations, use:

{
  "mcpServers": {
    "s3-mcp-server": {
      "command": "uv",
      "args": [
        "--directory",
        "PATH_TO_YOUR_PROJECT_DIRECTORY",
        "run",
        "s3-mcp-server"
      ]
    }
  }
}

Replace PATH_TO_YOUR_PROJECT_DIRECTORY with the actual path to your project directory.

Usage

Once installed and configured, you can use the AWS S3 Access server to interact with your S3 buckets. The server provides resources for retrieving PDF documents and tools for listing and accessing S3 objects.

Example Commands

To list all available buckets:

Use the ListBuckets tool to show me all my S3 buckets.

To list objects in a specific bucket:

Use the ListObjectsV2 tool to show me the contents of my "documents" bucket.

To retrieve and view a PDF document:

Use the GetObject tool to retrieve the file "reports/annual-2023.pdf" from my "company-docs" bucket.

Limitations

Currently only supports PDF documents
Limited to retrieving 1,000 objects at a time
Requires proper AWS credentials with appropriate permissions

Debugging

If you encounter issues with the MCP server, you can use the MCP Inspector for debugging:

Install the MCP Inspector using npm:

npx @modelcontextprotocol/inspector

Launch the inspector with your server:

npx @modelcontextprotocol/inspector uv --directory PATH_TO_YOUR_PROJECT_DIRECTORY run s3-mcp-server

The inspector will provide a URL that you can open in your browser to debug the server interactions.