The Data Quality Scores server imports and retrieves data quality scores for rule occurrences using Cloud Data Governance and Catalog (CDGC).
This MCP server gives AI agents a direct interface to data quality scores held in Data Governance and Catalog. It publishes computed rule-occurrence scores into the catalog and reads back their historic run values, letting agents ground quality assessments in governed, authoritative measurements. Trend and threshold analysis are composed by the agent over the returned run history.
Key capabilities
This MCP server offers the following key capabilities:
•Score publishing. Imports computed rule-occurrence runs and their quality scores into the CDGC catalog for governed storage.
•Quality trend analysis. Supplies the run-level scores agents need to compute trends and investigate quality changes over time.
•Historic score retrieval. Returns the run history for a rule occurrence, optionally filtered by date, source, or score threshold.
Use cases
Use this MCP server to address the following use cases:
•Quality monitoring. Agents check the current and historic scores of a rule occurrence to report on data health.
•Score publishing. Headless or scheduled jobs publish freshly computed quality scores back into the catalog.
•Trend investigation. Agents analyze score history over a date range to explain why an asset's quality changed.
Tools
The following table describes the tools available on this MCP server:
Tool
Description
get_dq_scores
Retrieves the historic run scores for a single data quality rule occurrence, optionally filtered by date, source, or score threshold.
import_dq_scores
Publishes data quality rule-occurrence runs with their computed scores into the CDGC catalog.
Note:
In AI Agent Engineering, the MCP server tools are called "actions."