MCP

Model Context Protocol lets you connect your Large Language Model (LLM) in, for example, your code editor to external data. Follow this guide to connect your editor to your Bucket data.

Get started with Bucket Remote MCP

The Model Context Protocol (MCP) is an open protocol that provides a standardized way to connect AI models to different data sources and tools. In the context of Bucket, MCP enables your code editor to understand your feature flags, their states, and their relationships within your codebase. This creates a seamless bridge between your feature management workflow and AI-powered development tools. Bucket hosts the MCP server, making it very easy to get started!

1

Set up the MCP connection with the Bucket CLI

Simply run our one-line command in your project directory and select the options that best suit you.

npx @bucketco/cli@latest mcp

2

That's it!

Make sure your editor/client and Node.js versions are up-to-date.

Manual setup

Below are the manual setup steps for some of the popular editors, but there are more MCP-compatible clients. Before you start with the manual setup, you first need the App ID from your app settings page.

We use mcp-remote to enable authentication and remote MCP connections in clients that don't yet support it.

Cursor

1. Open Cursor.

2. Go to Settings > MCP.

3. Click Add new global MCP server and paste the STDIO configure with your App ID:

{
  "mcpServers": {
    "Bucket": {
      "command": "npx",
      "args": [
        "mcp-remote@latest",
        "https://app.bucket.co/api/mcp?appId=<YOUR APP ID>"
      ]
    }
  }
}

Cursor also supports workspace-specific MCP servers by adding the above configuration to .cursor/mcp.json inside your project directory.

1. Save, go back to Cursor, and start prompting!

Visual Studio Code

You must enable the Copilot agent mode to use MCP in Visual Studio Code Copilot chat.

1. Open VS Code.

2. Open the command palette, typically CMD + SHIFT + P or CTRL + SHIFT + P.

3. Type and select MCP: Add Server... .

4. Select Command (stdio) .

5. Enter npx mcp-remote@latest https://app.bucket.co/api/mcp?appId=<YOUR APP ID> as the command, with your App ID.

6. Enter Bucket as the server ID.

7. Select either User Settings or Workspace Settings.

VS Code also supports workspace-specific MCP servers by adding the above configuration to .vscode/mcp.json inside your project directory.

1. Start prompting!

Claude Desktop

1. Open Cursor Desktop.

2. Go to Settings > Developer.

3. Click Edit Config , open claude_desktop_config.json and paste the STDIO configuration with your App ID:

{
  "mcpServers": {
    "Bucket": {
      "command": "npx",
      "args": [
        "mcp-remote@latest",
        "https://app.bucket.co/api/mcp?appId=<YOUR APP ID>"
      ]
    }
  }
}

1. Save, restart Claude Desktop, and start prompting!

Additional notes

MCP is in a very early stage of development, and changes are frequent. If something isn't working, please check out the Model Context Protocol Website, mcp-remote, and open an issue ticket here.