MCP Data Masking, Stewardship and Data Product Dashboards, Harvesting Semantic Links

Tenant controls to mask sensitive data sent to AI via MCP, richer responsibilities coverage and data product dashboards, and JDBC crawling that ingests semantic links to the business glossary.

This month brings stronger privacy controls for AI integrations, clearer stewardship analytics—including raw exports and gap views—and faster glossary alignment when importing technical metadata. The data products area now surfaces responsibility distribution alongside existing views, and JDBC-based crawling can carry semantic links into the catalog on ingest.

New Features and Improvements

  • MCP: configurable masking of sensitive data: Administrators can configure how content is handled before it is exposed to AI clients over MCP, so fields or patterns your organization does not want sent to external LLMs can be masked or withheld according to policy. This reduces the risk of leaking identifiers, or other sensitive metadata during natural-language discovery and assistant workflows. MCP Server , Connecting Your AI Client

  • Responsibilities coverage dashboard: The stewardship coverage experience now supports downloading raw underlying data for offline analysis and reporting, and a simplified gap analysis view that makes it easier to see where coverage falls short without wading through dense tables. Together, these updates help governance teams prioritize assignments and prove coverage to auditors. How to Monitor Stewardship Responsibilities

  • Data products dashboard—responsibilities tab: The data products dashboard includes a dedicated tab for responsibilities distribution. Use it to spot resources with missing or unassigned responsibilities and to monitor user workload across data products, so ownership gaps and overload show up in the same place as the rest of product-level governance. Data Products Catalog

  • JDBC crawling and semantic links to the glossary: JDBC metadata jobs can now ingest semantic links from data catalog assets to business glossary elements as part of the crawl. This is especially useful when another product or an AI-driven pipeline already knows which columns or tables map to glossary terms: you can bulk-populate those links on first import instead of reconciling them only inside Blindata afterward. Import Metadata with SQL or JDBC , Linking Catalog Assets to Business Glossary

Versions

  • blindata-api: v1.175.0
  • blindata-ui: v1.159.0