Unity Catalog MCP Server

What does “Unity Catalog” MCP Server do?

The Unity Catalog MCP Server serves as a Model Context Protocol (MCP) layer for the Unity Catalog, enabling AI assistants to interact programmatically with Unity Catalog’s data management functions. By exposing Unity Catalog Functions as MCP-compatible tools, this server allows developers and AI agents to perform operations like listing, querying, creating, and deleting database functions in a standardized way. This greatly enhances development workflows by automating data catalog tasks, facilitating seamless integration into flows and agents, and supporting dynamic access to schema and function metadata. The server is designed for environments where managing, discovering, and manipulating data functions programmatically is essential, such as in data engineering, analytics, and AI-augmented development.

List of Prompts

No prompt templates are listed in the repository or documentation.

List of Resources

No explicit MCP resources are documented in the repository or README.

List of Tools

• uc_list_functions
  • Lists functions within the specified parent catalog and schema. Returns a list of functions retrieved from Unity Catalog.
• uc_get_function
  • Retrieves details for a specific function within a parent catalog and schema. Input: name (string).
• uc_create_function
  • Creates a new function in the specified catalog and schema. Input: name (string), script (string). Experimental API.
• uc_delete_function
  • Deletes a function in the specified catalog and schema. Input: name (string).

Additionally, all Unity Catalog Functions registered in Unity Catalog are available as tools.

Use Cases of this MCP Server

• Automated Function Management
  • Enables AI or developers to list, create, and delete catalog functions programmatically, streamlining data engineering operations.
• Catalog and Schema Exploration
  • Facilitates exploration and documentation of functions within Unity Catalog for data discovery and auditing.
• Programmatic Function Registration
  • Supports registering new Python-based functions into Unity Catalog directly from workflows or AI agents.
• Access Control and Security Automation
  • Allows secure, token-based interaction with catalog resources, fitting into automated compliance or governance workflows.

How to set it up

Windsurf

No platform-specific instructions available for Windsurf.

Claude

1. Ensure prerequisites such as uv or Python are installed.
2. Obtain the Unity Catalog server URL, catalog, schema, and (optionally) an access token.
3. Add the Unity Catalog MCP Server to your claude_desktop_config.json:

   {
     "mcpServers": {
       "unity-catalog": {
         "command": "uv",
         "args": [
           "run",
           "mcp-server-unitycatalog",
           "--uc_server", "<UNITY_CATALOG_URL>",
           "--uc_catalog", "<CATALOG_NAME>",
           "--uc_schema", "<SCHEMA_NAME>",
           "--uc_token", "${UC_TOKEN}"
         ]
       }
     }
   }
   

4. Save the configuration and restart Claude Desktop.
5. Verify connection in the MCP tool section.

Securing API Keys Example:

{
  "mcpServers": {
    "unity-catalog": {
      "command": "uv",
      "args": [ "run", "mcp-server-unitycatalog", "--uc_server", "...", "--uc_token", "${UC_TOKEN}" ],
      "env": {
        "UC_TOKEN": "your_secret_token"
      },
      "inputs": {
        "UC_TOKEN": "env"
      }
    }
  }
}

Cursor

No platform-specific instructions available for Cursor.

Cline

1. Ensure you have installed the necessary dependencies and uv.
2. Configure as in Claude, using your cline_config.json:

   {
     "mcpServers": {
       "unity-catalog": {
         "command": "uv",
         "args": [
           "run",
           "mcp-server-unitycatalog",
           "--uc_server", "<UNITY_CATALOG_URL>",
           "--uc_catalog", "<CATALOG_NAME>",
           "--uc_schema", "<SCHEMA_NAME>",
           "--uc_token", "${UC_TOKEN}"
         ]
       }
     }
   }
   

3. Save and restart Cline.
4. Check that the MCP server appears under available tools.

Securing API Keys Example: (see previous)

How to use this MCP inside flows

Using MCP in FlowHunt

To integrate MCP servers into your FlowHunt workflow, start by adding the MCP component to your flow and connecting it to your AI agent:

Click on the MCP component to open the configuration panel. In the system MCP configuration section, insert your MCP server details using this JSON format:

{
  "unity-catalog": {
    "transport": "streamable_http",
    "url": "https://yourmcpserver.example/pathtothemcp/url"
  }
}

Once configured, the AI agent is now able to use this MCP as a tool with access to all its functions and capabilities. Remember to change "unity-catalog" to whatever the actual name of your MCP server is and replace the URL with your own MCP server URL.

Overview

Our opinion

The Unity Catalog MCP Server is well-documented in terms of its tools and setup for Claude and Cline platforms, with a clear license and basic security (API key) instructions. However, it lacks explicit documentation for prompt templates and MCP resources, and there’s no mention of advanced MCP features such as roots or sampling. Overall, it’s a solid, focused integration for Unity Catalog but could be improved with additional documentation and MCP feature support.

MCP Score