genai-toolbox/docs/DATAPLEX_README.md

Dataplex MCP Server

The Dataplex Model Context Protocol (MCP) Server gives AI-powered development tools the ability to work with your Google Cloud Dataplex Catalog. It supports searching and looking up entries and aspect types.

Features

An editor configured to use the Dataplex MCP server can use its AI capabilities to help you:

• Search Catalog - Search for entries in Dataplex Catalog
• Explore Metadata - Lookup specific entries and search aspect types

Installation and Setup

Prerequisites

• Download and install MCP Toolbox:

  1. Download the Toolbox binary: Download the latest binary for your operating system and architecture from the storage bucket. Check the releases page for OS and CPU architecture support: https://storage.googleapis.com/genai-toolbox/v0.21.0/<os>/<arch>/toolbox

     • Replace <os> with linux, darwin (macOS), or windows.
     • Replace <arch> with amd64 (Intel) or arm64 (Apple Silicon).

     curl -L -o toolbox https://storage.googleapis.com/genai-toolbox/v0.21.0/linux/amd64/toolbox
     

  2. Make it executable:

     chmod +x toolbox
     

  3. Move binary to /usr/local/bin/ or /usr/bin/:

     sudo mv toolbox /usr/local/bin/
     # sudo mv toolbox /usr/bin/
     

     On Windows, move binary to the WindowsApps\ folder:

     move "C:\Users\<path-to-binary>\toolbox.exe" "C:\Users\<username>\AppData\Local\Microsoft\WindowsApps\"
     

     Tip: Ensure the destination folder for your binary is included in your system's PATH environment variable. To check PATH, use echo $PATH (or echo %PATH% on Windows).

     Note: You may need to restart Antigravity for changes to take effect.

• A Google Cloud project with the Dataplex API enabled.

• Ensure Application Default Credentials are available in your environment.

• IAM Permissions:

  • Dataplex Viewer (roles/dataplex.viewer) or equivalent permissions to read catalog entries.

Configuration

The MCP server is configured using environment variables.

export DATAPLEX_PROJECT="<your-gcp-project-id>"

Add the following configuration to your MCP client (e.g., settings.json for Gemini CLI):

{
  "mcpServers": {
    "dataplex": {
      "command": "toolbox",
      "args": ["--prebuilt", "dataplex", "--stdio"],
      "env": {
        "DATAPLEX_PROJECT": "your-project-id"
      }
    }
  }
}

Usage

Once configured, the MCP server will automatically provide Dataplex capabilities to your AI assistant. You can:

• "Search for entries related to 'sales' in Dataplex."
• "Look up details for the entry 'projects/my-project/locations/us-central1/entryGroups/my-group/entries/my-entry'."

Server Capabilities

The Dataplex MCP server provides the following tools:

Tool Name	Description
search_entries	Search for entries in Dataplex Catalog.
lookup_entry	Retrieve a specific entry from Dataplex Catalog.
search_aspect_types	Find aspect types relevant to the query.

Documentation

For more information, visit the Dataplex documentation.