AWS S3 MCP by Samuraikun is a Model Context Protocol server built with TypeScript and Node.js. It connects AI assistants to Amazon S3, enabling interaction with cloud storage. Distributed as an npm package and supporting containerized deployment, it provides AI agents with direct access to S3 buckets. This tool allows AI clients like Claude and Amazon Q Developer to manage and analyze S3-stored data. Its design focuses on integrating S3 capabilities into AI agent pipelines, making cloud storage operations accessible through natural language commands for various AI applications.

What does AWS S3 MCP connect to?

This MCP server establishes connections to Amazon S3 (Simple Storage Service). It functions as a Model Context Protocol server, specifically designed to allow AI assistants to interact with external systems. Its core purpose is to bridge AI capabilities with S3 cloud storage, enabling AI agents to perform operations directly on S3 buckets. This integration makes S3 resources available for AI-driven processes, extending the reach of AI applications into cloud data management.

Core features and tools

AWS S3 MCP exposes specific callable tools for AI agents. These include list-buckets to enumerate available S3 buckets, list-objects for viewing contents within a specified bucket, and get-object to retrieve individual files. It supports STDIO Transport for direct process communication, HTTP Transport (REST API with Server-Sent Events) for web clients, and Streamable HTTP for real-time data streaming. Compatible AI clients confirmed are Claude, Cursor, Claude Desktop, Amazon Q Developer, and Cline.

Who is it for?

AWS S3 MCP targets developers and data engineers who build AI agent pipelines and Large Language Models. It is valuable for those needing to integrate S3 operations directly into AI workflows, allowing AI agents to manage and interact with cloud storage. This tool supports the creation of AI applications that require data persistence, retrieval, and analysis from S3. Skip this if your AI applications do not require direct interaction with Amazon S3.

How to get started

The server authenticates using standard AWS SDK credential resolution. This process prioritizes environment variables such as AWS_ACCESS_KEY_ID and AWS_SECRET_ACCESS_KEY, followed by a shared credentials file (~/.aws/credentials), or EC2 Instance Metadata if running on an EC2 instance with an IAM role. Accessible S3 buckets must be explicitly specified via the S3_BUCKETS environment variable for the server to operate correctly.

Use cases

• Fetching documents for Retrieval Augmented Generation (RAG).
• Persisting outputs from AI agents to cloud storage.
• Dynamically managing S3 files through natural language commands.
• Enabling AI agents to read and analyze S3 file contents, including built-in PDF parsing.

AWS S3 MCP by Samuraikun offers a direct pathway for AI agents to interact with Amazon S3, supporting various data management and analysis tasks. Its design with multiple transport options and standard AWS authentication makes it suitable for integrating cloud storage into AI pipelines. The tool empowers AI agents to perform S3 operations, from listing contents to retrieving objects. Try enabling an AI agent to read and analyze the contents of a PDF file stored in S3.