S3 Mcp

What is S3 Mcp?

S3 Mcp is a Model Context Protocol server that enables AI assistants to manage Amazon S3 buckets and objects directly. It provides tools for bucket and object operations, lifecycle policies, tagging, CORS configuration, and more. It is intended for developers who want to integrate S3 management into AI-driven workflows.

How to use S3 Mcp?

S3 Mcp can be run via Docker with AWS credentials passed as environment variables. It can be configured in Cursor by adding a Docker-based MCP server entry to the Cursor MCP configuration. Manual Docker execution is also supported, and the server can be built from the repository.

Key features of S3 Mcp

• List, create, and delete S3 buckets
• List, get, put, delete, and copy objects
• Generate presigned URLs for objects
• Manage bucket policies (set, get, delete)
• Configure lifecycle rules and CORS
• Set and retrieve object tags
• Upload local files and download to local path

Use cases of S3 Mcp

• List all S3 buckets from an AI chat interface
• Create a new bucket with versioning and encryption enabled
• Upload a local file to an S3 bucket
• Retrieve object tags for a specific file
• Generate a presigned URL for temporary object access

FAQ from S3 Mcp

What runtime or dependencies are required?

Docker is required to run the pre-built image or to build the server from source. AWS credentials must be provided via environment variables.

How are AWS credentials supplied?

Credentials are passed as AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY, and AWS_REGION environment variables when running the container.

Where does the data live?

All data operations are performed directly on Amazon S3. The server itself does not store any persistent data.

What transports are supported?

The server uses standard input/output (stdio) communication, as typical for MCP servers run via Docker.

Is the server authenticated?

Authentication is handled via the provided AWS credentials, not through a separate MCP-level auth mechanism.