Minio MCP Service

@ucesys

About Minio MCP Service

Minio MCP Python Implementation

Config

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

{
  "mcpServers": {
    "minio_service": {
      "command": "/Users/username/.local/bin/uv",
      "args": [
        "--directory",
        "/Users/username/PycharmProjects/minio-mcp/src/minio_mcp_server",
        "run",
        "server.py"
      ]
    }
  }
}

Tools

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

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

Retrieves an object from MinIO

Uploads a file to MinIO bucket using fput method

Overview

What is Minio MCP Service?

This project implements a Model Context Protocol (MCP) server and client for MinIO object storage, providing a standardized way to interact with MinIO.

How to use Minio MCP Service?

Install dependencies using pip or uv, configure a .env file with MinIO and server settings, then run the server with python src/minio_mcp_server/server.py. Use the basic client via src/client.py or the Anthropic client with src/client/mcp_anthropic_client.py.

Key features of Minio MCP Service

Exposes MinIO data through Resources (text and binary files)
Provides Tools: ListBuckets, ListObjects, GetObject, PutObject
Supports pagination and filtering options on ListBuckets and ListObjects
Includes basic and Anthropic-powered client implementations
Integrates with Claude Desktop via stdio MCP server

Use cases of Minio MCP Service

Programmatic access to MinIO object storage through a standard MCP interface
AI-powered queries and file operations using Anthropic’s Claude models
Integration of MinIO data into Claude Desktop workflows
Automated file uploads, listing, and retrieval from MinIO buckets

FAQ from Minio MCP Service

What dependencies are needed?

Python and the packages in requirements.txt. Install with pip install -r requirements.txt or uv pip install -r requirements.txt.

How do I configure the server?

Create a .env file in the root with MinIO endpoint, access/secret keys, security flag, and optional server host/port.

What are the known limits?

ListObjects returns up to 1,000 objects per request. ListBuckets supports pagination via start_after and max_buckets parameters. Bucket contents are limited to 1000 objects.

How does the server communicate?

The MCP server runs over stdio. Debugging can be done using the MCP Inspector: npx @modelcontextprotocol/inspector python path/to/server.py.

Can I use this with Claude Desktop?

Yes. Add a server entry to the Claude Desktop config file pointing to the server script with command: "python" and args including the server path.

Frequently asked questions

What dependencies are needed?

Python and the packages in `requirements.txt`. Install with `pip install -r requirements.txt` or `uv pip install -r requirements.txt`.

How do I configure the server?

Create a `.env` file in the root with MinIO endpoint, access/secret keys, security flag, and optional server host/port.

What are the known limits?

ListObjects returns up to 1,000 objects per request. ListBuckets supports pagination via `start_after` and `max_buckets` parameters. Bucket contents are limited to 1000 objects.

How does the server communicate?

The MCP server runs over stdio. Debugging can be done using the MCP Inspector: `npx @modelcontextprotocol/inspector python path/to/server.py`.

Can I use this with Claude Desktop?

Yes. Add a server entry to the Claude Desktop config file pointing to the server script with `command: "python"` and `args` including the server path.