Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Data Product MCP: Chat with your Enterprise Data

Marco Tonnarelli, Filippo Scaramuzza, Simon Harrer, Linus W. Dietz

TL;DR

The paper addresses the governance-friction barrier impeding enterprise AI by proposing Data Product MCP, a chat-driven, governance-preserving data access system built on the Model Context Protocol and a data product marketplace. It integrates semantic discovery, purpose-based access control, and real-time contract enforcement across federated data sources, enabling AI agents to discover, request, and query data with auditable governance, including an immutable audit trail. The implementation comprises five components (Data Product Marketplace, Governance AI, Data Sources, Data Product MCP Server, and an Agent chat interface) and four MCP-enabled tools, validated through qualitative feedback from $n=16$ domain experts across data governance roles. Results indicate improved natural-language data access, stronger governance, and faster self-service, while highlighting challenges in metadata quality, security, scalability, and trust, and pointing toward practical impact for Data Mesh adoption.

Abstract

Computational data governance aims to make the enforcement of governance policies and legal obligations more efficient and reliable. Recent advances in natural language processing and agentic AI offer ways to improve how organizations share and use data. But many barriers remain. Today's tools require technical skills and multiple roles to discover, request, and query data. Automating data access using enterprise AI agents is limited by the means to discover and autonomously access distributed data. Current solutions either compromise governance or break agentic workflows through manual approvals. To close this gap, we introduce Data Product MCP integrated in a data product marketplace. This data marketplace, already in use at large enterprises, enables AI agents to find, request, and query enterprise data products while enforcing data contracts in real time without lowering governance standards. The system is built on the Model Context Protocol (MCP) and links the AI-driven marketplace with cloud platforms such as Snowflake, Databricks, and Google Cloud Platform. It supports semantic discovery of data products based on business context, automates access control by validating generated queries against approved business purposes using AI-driven checks, and enforces contracts in real time by blocking unauthorized queries before they run. We assessed the system with feedback from $n=16$ experts in data governance. Our qualitative evaluation demonstrates effectiveness through enterprise scenarios such as customer analytics. The findings suggest that Data Product MCP reduces the technical burden for data analysis without weakening governance, filling a key gap in enterprise AI adoption.

Data Product MCP: Chat with your Enterprise Data

TL;DR

The paper addresses the governance-friction barrier impeding enterprise AI by proposing Data Product MCP, a chat-driven, governance-preserving data access system built on the Model Context Protocol and a data product marketplace. It integrates semantic discovery, purpose-based access control, and real-time contract enforcement across federated data sources, enabling AI agents to discover, request, and query data with auditable governance, including an immutable audit trail. The implementation comprises five components (Data Product Marketplace, Governance AI, Data Sources, Data Product MCP Server, and an Agent chat interface) and four MCP-enabled tools, validated through qualitative feedback from domain experts across data governance roles. Results indicate improved natural-language data access, stronger governance, and faster self-service, while highlighting challenges in metadata quality, security, scalability, and trust, and pointing toward practical impact for Data Mesh adoption.

Abstract

Computational data governance aims to make the enforcement of governance policies and legal obligations more efficient and reliable. Recent advances in natural language processing and agentic AI offer ways to improve how organizations share and use data. But many barriers remain. Today's tools require technical skills and multiple roles to discover, request, and query data. Automating data access using enterprise AI agents is limited by the means to discover and autonomously access distributed data. Current solutions either compromise governance or break agentic workflows through manual approvals. To close this gap, we introduce Data Product MCP integrated in a data product marketplace. This data marketplace, already in use at large enterprises, enables AI agents to find, request, and query enterprise data products while enforcing data contracts in real time without lowering governance standards. The system is built on the Model Context Protocol (MCP) and links the AI-driven marketplace with cloud platforms such as Snowflake, Databricks, and Google Cloud Platform. It supports semantic discovery of data products based on business context, automates access control by validating generated queries against approved business purposes using AI-driven checks, and enforces contracts in real time by blocking unauthorized queries before they run. We assessed the system with feedback from experts in data governance. Our qualitative evaluation demonstrates effectiveness through enterprise scenarios such as customer analytics. The findings suggest that Data Product MCP reduces the technical burden for data analysis without weakening governance, filling a key gap in enterprise AI adoption.
Paper Structure (19 sections, 3 figures)

This paper contains 19 sections, 3 figures.

Table of Contents

  1. Introduction
  2. Related Work
  3. System Architecture
  4. Data Product Marketplace
  5. Governance AI
  6. Data Sources
  7. Data Product MCP Server
  8. Data Product Search
  9. Data Product Get
  10. Data Product Request Access
  11. Data Product Query
  12. Agent
  13. Scenario-based Demonstration of Data Product MCP
  14. Identifying Top Customers with the Human in the Loop
  15. Analyzing Support Cases Without Additional Human Oversight
  16. ...and 4 more sections

Figures (3)

  • Figure 1: System architecture for AI-driven data product access and governance. The user formulates a business query, for which the Data Product MCP server identifies suitable data sources. Data access is then requested specifically for this query, following governance policies that determine whether approval is automatic or requires review by data owners. Once approved, the MCP server grants controlled access, formulates, and executes the corresponding data queries. The Data Product Marketplace functions as a centralized exchange for data products, ensuring compliance, traceability, and user accountability throughout the entire data lifecycle.
  • Figure 2: Identifying Top Customers with the Human in the Loop
  • Figure 3: Preventing Misuse of Customer Data at Query Time