🦁 Tigris MCP Server

[!IMPORTANT] Tigris has a hosted MCP server with OAuth support. Read more about it on mcp.storage.dev.

Tigris is a high-performance, S3-compatible object storage system designed for multi-cloud and AI workloads. We move your data all around the world based on where it's needed so that downloads are fast and the data is close to your users. You can store anything you want on Tigris (AI models, training data, database backups, request logs, social media uploads, or anything else) with no egress fees.

The Tigris MCP Server implements the MCP specification to create a seamless connection between AI agents and Tigris key features like bucket and object management.

🎯 Features

The Tigris MCP server provides your agents context to your Tigris buckets and objects. That allows you to use Tigris in your AI editor workflows.

Here are some example prompts you can try:

📦 Buckets

• List my tigris buckets
• Create a new tigris bucket and call it my-bucket
• Delete my tigris bucket called my-bucket

🔗 Objects

• List my objects in bucket my-bucket
• Upload /Users/ME/tigris-mcp-server/myfile.txt to my-bucket
• Create a folder named test in my-bucket
• Create a file test.txt with content Hello World in my-bucket
• Give me a link to share for test.txt
• Delete myfile.txt from my-bucket

Checkout our blog post about Vibe coding with Tigris MCP Server and more tips on sharing files using Tigris MCP Server

🚀 Getting Started

As Tigris supports the S3 API, you can use the wide range of available S3 tools, libraries, and extensions. You can get AWS_ACCESS_KEY_ID and AWS_SECRET_ACCESS_KEY from webconsole by following these the steps. Please refer to our Tigris Data documentation for detailed overview.

To get started:

• Sign up for an account at storage.new.
• Get an access key at storage.new/accesskey.
• Copy the AWS_ACCESS_KEY_ID and AWS_SECRET_ACCESS_KEY to a safe place like your password manager. These will not be shown again.

⚒️ Requirements

Tigris MCP server can be used both with npx and docker. We recommend running with docker as it provides better sandboxing.

We support installing the Tigris MCP server two ways:

1. 🐳 Docker
2. 📦 NPX

We suggest installing and using the MCP server with Docker as it provides much better sandboxing than NPX.

• Running the Tigris MCP server with docker requires the Docker Engine to be installed. If you don't have it installed, follow the instructions here.
• Running the Tigris MCP server with npx requires Node.js to be installed. If you don't have it installed, follow the instructions here.

⚙️ Installation

🪄 One click install for VS Code

Click one of these buttons to install the Tigris MCP Server for VS Code or VS Code Insiders.

📦 Claude Desktop and Cursor AI

You can install the Tigris MCP server in Claude Desktop and Cursor by running our install script:

Copy code
npx -y @tigrisdata/tigris-mcp-server init

🪏 Manual Installation

If you don't want to use our automatic install script, you can manually install the Tigris MCP server by adding one of these blocks to your MCP client's configuration.

For Claude Desktop, edit one of the following files:

• 🍎 macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
• 🪟 Windows: %APPDATA%\Claude\claude_desktop_config.json

To open the right file in Cursor:

• 🍎 macOS:
  • Open the Cursor menu in the upper-left hand corner of your screen
  • Go to Settings -> Cursor Settings
  • Click on the MCP tab
  • Click "Add new global MCP server"
• 🪟 Windows:
  • Open the File menu
  • Go to Preferences -> Cursor Settings
  • Click on the MCP tab
  • Click "Add new global MCP server"

Then add one of the following blocks to the end of your configuration:

📦 Via NPX

jsonCopy code
{
  "mcpServers": {
    "tigris-mcp-server": {
      "command": "npx",
      "args": ["-y", "@tigrisdata/tigris-mcp-server", "run"],
      "env": {
        "AWS_ACCESS_KEY_ID": "YOUR_AWS_ACCESS_KEY_ID",
        "AWS_SECRET_ACCESS_KEY": "YOUR_AWS_SECRET_ACCESS_KEY",
        "AWS_ENDPOINT_URL_S3": "https://fly.storage.tigris.dev"
      }
    }
  }
}

🐳 Via Docker

Please note that the server will only allow operations within `/User/CURRENT_USER/tigris-mcp-server. This allows for a secure sandboxing environment.

jsonCopy code
{
  "mcpServers": {
    "tigris-mcp-server": {
      "command": "docker",
      "args": [
        "-e",
        "AWS_ACCESS_KEY_ID",
        "-e",
        "AWS_SECRET_ACCESS_KEY",
        "-e",
        "AWS_ENDPOINT_URL_S3",
        "-i",
        "--rm",
        "--mount",
        "type=bind,src=/Users/CURRENT_USER/tigris-mcp-server,dst=/Users/CURRENT_USER/tigris-mcp-server",
        "quay.io/tigrisdata/tigris-mcp-server:latest"
      ],
      "env": {
        "AWS_ACCESS_KEY_ID": "YOUR_AWS_ACCESS_KEY_ID",
        "AWS_SECRET_ACCESS_KEY": "YOUR_AWS_SECRET_ACCESS_KEY",
        "AWS_ENDPOINT_URL_S3": "https://fly.storage.tigris.dev"
      }
    }
  }
}

Alternatively, you can use your existing AWS Profiles if you have AWS CLI installed and have your AWS credential configured. You can use the following configuration.

jsonCopy code
{
  "mcpServers": {
    "tigris-mcp-server": {
      "command": "npx",
      "args": ["-y", "@tigrisdata/tigris-mcp-server", "run"],
      "env": {
        "USE_AWS_PROFILES": "true",
        "AWS_PROFILE": "default",
        "AWS_ENDPOINT_URL_S3": "https://fly.storage.tigris.dev"
      }
    }
  }
}

or via docker

jsonCopy code
{
  "mcpServers": {
    "tigris-mcp-server": {
      "command": "docker",
      "args": [
        "run",
        "-e",
        "USE_AWS_PROFILES",
        "-e",
        "AWS_PROFILE",
        "-e",
        "AWS_ENDPOINT_URL_S3",
        "-i",
        "--rm",
        "--mount",
        "type=bind,src=/Users/CURRENT_USER/tigris-mcp-server,dst=/Users/CURRENT_USER/tigris-mcp-server",
        "quay.io/tigrisdata/tigris-mcp-server:latest"
      ],
      "env": {
        "USE_AWS_PROFILES": "true",
        "AWS_PROFILE": "default",
        "AWS_ENDPOINT_URL_S3": "https://fly.storage.tigris.dev"
      }
    }
  }
}

For development, refer to the CONTRIBUTING.md file.