RELATED 01
dbt MCP Server
OFFICIAL SERVERdbt Labs' server for running dbt commands, querying metrics and reading project metadata
617GitHub stars
76k/moPyPI downloads
81popularity score
OpenMetadata MCP Server
Lets an assistant search a data catalog, trace lineage and update metadata and tests in OpenMetadata.
Official · OpenMetadata Java Data and analyticsDatabases
18 popularity score ★ 16k
What it does
OpenMetadata is an open-source metadata platform: a data catalog that connects technical metadata, lineage, data quality, ownership, glossaries, policies and team knowledge across an organisation's data. It includes an MCP server so assistants and agents can work with that metadata graph in natural language.
An assistant can search assets and business terms by keyword or by meaning, read full entity details, see who owns a dataset and what a metric means, trace upstream and downstream lineage, run root-cause analysis on data quality failures, and read company context notes and persona context. With write access it can create entities and lineage, update descriptions, tags and owners with JSON Patch, and list and create data quality test cases.
The server is part of an OpenMetadata deployment (self-hosted or managed), not a separate package. Interactive clients such as Claude Desktop, Cursor and VS Code sign in with OAuth using the same login as the OpenMetadata UI; agents can use a personal access token or, for unattended use, a bot token sent as a Bearer token.
Tools 15
search_metadata- Find data assets and business terms by keyword: names, owners, tags, tiers, services or column names.
semantic_search- Find data assets by meaning using vector search.
company_context- Read Company Context knowledge pills (question and answer notes from files in the Context Center).
get_entity_details- Retrieve full details for an entity by fully qualified name.
get_user_context- Return context about the authenticated user: identity, teams, roles, domains and persona.
get_persona_context- Return the shared AI context document curated for a persona.
find_context- Semantically search the company-knowledge layer and return definitions plus candidate data assets.
get_entity_lineage- Retrieve upstream and downstream lineage for an entity for impact analysis.
create_lineage- Create a lineage relationship between two entities.
root_cause_analysis- Trace a data quality failure back to its origin across pipeline hops.
create_entity- Create a new entity of a given type.
describe_entity_type- Return the attributes create_entity accepts for an entity type.
patch_entity- Update an existing entity's properties with JSON Patch.
get_test_definitions- List data quality test definitions for tables or columns.
create_test_case- Create a data quality test case for a table or column.