Getting started with watsonx.data intelligence MCP server

Securely access, explore your data, and complete tasks across data governance and catalog, data quality, data lineage, and Data Product Hub through natural language by using IBM watsonx.data intelligence Model Context Protocol (MCP) server and an AI agent.

Connection models

Watsonx.data intelligence supports two MCP server connection models. You can choose the type of MCP server that best fits your needs.

Remote MCP server (managed)
The remote MCP server is a hosted endpoint on IBM Cloud or AWS.
Local MCP server
The local MCP server runs on your local machine.

Configuration workflow

The configuration process varies depending on whether you choose the remote or local MCP server connection model. Follow the workflow that matches your chosen connection model.

For remote MCP server

To connect managed watsonx.data intelligence MCP server with an AI agent, complete these main tasks:

  1. Configure the client to authenticate by using your API credentials. For more information, see Managing the user API key. Authentication is required to ensure that the tool access is authorized and scoped to your user or service identity.

  2. Register the managed MCP server as a remote MCP server in the configuration of your AI agent.
    You can configure an AI agent. This list includes, but is not limited to the following AI agents:

    For other AI agents, you can configure the MCP server by using environment variables or an .env file. For more information, see MCP configuration file and environment variables.

  3. Optional: Set up skills as described in Setting up and using skills.

For local MCP server

  1. Install the MCP server on your local machine.

  2. Register the local MCP server in the configuration of an AI agent of your choice.
    You can configure any AI agent that supports MCP integrations. This list includes, but is not limited to the following AI agents:

    For other AI agents, you can configure the MCP server by using environment variables or an .env file. For more information, see MCP configuration file and environment variables.

  3. Optional: Set up skills as described in Setting up and using skills.

Capabilities

Ask your AI agent to list the tasks that it can help you complete.

Tools

You can use tools across the following areas to complete tasks:

  • Connection management
  • Data product management
  • Data protection and governance
  • Data quality
  • Lineage
  • Semantic search and query
  • Text-to-SQL
  • Metadata enrichment
  • Metadata import
  • Project and container management
  • Search and discovery
  • Reporting
  • Workflow management

For a detailed tools list with descriptions and prompts, see MCP Tools Reference.

Skills

Skills are modular, reusable capabilities that provide AI agents with specialized instructions how to perform specific tasks. To use skills, you must download the skills folder and copy its contents to a location based on the requirements of your AI agent as described in Setting up and using skills.

For a detailed skills list with descriptions and prompts, see MCP Skills.

Understanding the interaction model

The MCP server enables conversational data access through the following workflow:

  • You submit a natural language query to your AI agent.
  • The AI agent interprets your request and selects the appropriate MCP tool.
  • The MCP server executes the operation against your watsonx.data intelligence instance.
  • Results are returned to the AI agent.
  • The AI agent presents the results in a conversational format.