OpenMetadata

Overview

OpenMetadata is a comprehensive metadata platform that combines technical metadata, data quality signals, lineage, and business semantics into a unified knowledge graph. Its MCP server enables AI assistants to interact with metadata through natural language, allowing for search, entity retrieval, lineage inspection, and metadata updates. The platform supports 120+ data connectors and provides contextual understanding of data assets including ownership, usage patterns, quality metrics, and business definitions.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

• search_metadata

  Search metadata in the knowledge graph using natural language queries

  Try:Find datasets related to customer behaviorTry:Show me tables containing PII data
• semantic_search

  Search by meaning rather than exact keywords

  Try:Find assets related to revenue forecastingTry:Show dashboards for user engagement metrics
• retrieve_entity_details

  Retrieve detailed information about a specific metadata entity

  Try:Get details for the customer_orders tableTry:Show information for the monthly_revenue metric
• inspect_lineage

  Examine upstream and downstream data lineage relationships

  Try:Show what depends on the customer_id columnTry:Trace where the transactions table flows to
• create_glossary

  Create a new glossary in the metadata system

  Try:Create a new product glossaryTry:Set up a financial terms glossary
• create_glossary_term

  Add a new term to an existing glossary

  Try:Add lifetime value to the product glossaryTry:Define churn in the business metrics glossary
• create_lineage

  Create lineage relationships between data assets

  Try:Create lineage between source and staging tablesTry:Show how the ETL pipeline transforms data
• update_metadata

  Update descriptions, tags, owners, and other metadata attributes

  Try:Update the description of the sales tableTry:Mark customer data as sensitive
• list_data_quality_tests

  List available data quality test definitions

  Try:Show all data quality tests for the customer tableTry:List freshness checks for inventory data
• create_data_quality_test

  Create new data quality test cases

  Try:Create a uniqueness test for email addressesTry:Add a freshness check for daily reports
• analyze_quality_failures

  Analyze root causes of data quality failures

  Try:Analyze why the data quality test failedTry:Investigate the pipeline failure for customer data
• get_entity_owners

  Retrieve ownership information for metadata entities

  Try:Who owns the customer data product?Try:Show me the owners of the marketing dashboard

Comparable tools

Installation

{
  "mcpServers": {
    "open-metadata": {
      "command": "npx",
      "args": ["@open-metadata/mcp-server", "--port", "8090"],
      "env": {
        "OPENMETADATA_SERVER_URL": "http://localhost:8585/api"
      }
    }
  }
}

FAQ

What tools does the OpenMetadata MCP server provide?

The MCP server provides tools for metadata search, semantic search, entity retrieval, lineage inspection, glossary management, lineage creation, metadata updates, and data quality test management.

How does OpenMetadata help AI systems understand data?

OpenMetadata provides comprehensive context including technical metadata, data quality signals, lineage, business semantics, ownership information, and governance policies that AI systems need to understand and trust data assets.

On Hacker News

Recent discussion from the developer community.

• OpenMetadata: Open Standard for Metadata▲ 4
  Story by mithunmanohar79 · 2022-05-24

Compare OpenMetadata with