# Local MCP Server

note

Please prefer using the [Remote MCP server whenever possible](/docs/mcp/remote/.md). Using the local MCP server employs a significantly different threat model than the remote MCP server. Please consult with your local systems administrator if you need a better understanding of the risks involved.

The **Tigris MCP Server** implements the [MCP specification](https://modelcontextprotocol.io/) to create a seamless connection between AI agents and Tigris key features like bucket and object management.

## Features[​](#features "Direct link to 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[​](#buckets "Direct link to Buckets")

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

### Objects[​](#objects "Direct link to 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](https://www.tigrisdata.com/blog/vibes-off/) with Tigris MCP Server and more tips on [sharing files](https://www.tigrisdata.com/blog/mcp-server-sharing/) using the Tigris MCP Server.

## Installation[​](#installation "Direct link to Installation")

As every tool has its own setup flows for MCP servers, each will be explained in detail in its own subsection:

### Claude Desktop[​](#claude-desktop "Direct link to Claude Desktop")

Run [our setup script](https://github.com/tigrisdata/tigris-mcp-server/blob/main/src/init.ts) to automate setup:

```
npx -y @tigrisdata/tigris-mcp-server init
```

### Cursor AI[​](#cursor-ai "Direct link to Cursor AI")

Run [our setup script](https://github.com/tigrisdata/tigris-mcp-server/blob/main/src/init.ts) to automate setup:

```
npx -y @tigrisdata/tigris-mcp-server init
```

### OpenAI Codex[​](#openai-codex "Direct link to OpenAI Codex")

Run `codex mcp add` to add the Tigris MCP server:

```
codex mcp add tigris --env AWS_PROFILE=tigris -- npx -y @tigrisdata/tigris-mcp-server run
```

Be sure that your credentials are configured [in your AWS configuration](/docs/sdks/s3/aws-cli/.md#setting-the-endpoint-url-in-credentials-file) and replace the AWS profile name as appropriate.

### VS Code (and VS Code Insiders)[​](#vs-code-and-vs-code-insiders "Direct link to VS Code (and VS Code Insiders)")

Run [our setup script](https://github.com/tigrisdata/tigris-mcp-server/blob/main/src/init.ts) to automate setup:

```
npx -y @tigrisdata/tigris-mcp-server init
```

Or do the following:

1. Press Control+Shift+P or Command+Shift+P to open the command palette.
2. Search for "MCP: Add Server".
3. Create a Command (stdio) MCP server.
4. Enter the command `npx -y @tigrisdata/tigris-mcp-server run`.
5. Name the MCP server `tigris`.
6. Authenticate with Tigris and get off to work!