Skip to content
Data MCP Directory

SIDE BY SIDE, read 10 Nov 2026

DataHub MCP Server and OpenMetadata MCP Server, field by field: 22 and 15 tools named

DataHub MCP ServerOpenMetadata MCP Server

The same fourteen fields for both servers, read 10 Nov 2026.
FieldDataHub MCP ServerOpenMetadata MCP Server
Data toolDataHubOpenMetadata
PublisherAcryl Data (DataHub)OpenMetadata
KindOfficialOfficial
Repository or docsgithub.com/acryldata/mcp-server-datahubdocs.open-metadata.org/v2.0.x/how-to-guides/mcp.md
Version or commit0.8.02.0.7 (bundled with platform release 2.0.7-release; 2.1.0-rc1 also listed)
Last release21 Oct 202627 Oct 2026
TransportRemote and localRemote (hosted)
AuthStdio: DATAHUB_GMS_URL + DATAHUB_GMS_TOKEN (or ~/.datahubenv); HTTP: per-request DataHub personal access token as Authorization Bearer, no shared server token.OAuth 2.0 via existing SSO (per user), personal access token, or Bot JWT as Authorization Bearer.
AccessConfigurableRead and write
Tools listed22 named in the evidence15 named in the evidence
Tool definitions read atgithub.com/acryldata/mcp-server-datahub/blob/main/README.mddocs.open-metadata.org/v2.0.x/how-to-guides/mcp/reference.md
Context weight567 tokens for 22 tool definitions (22 with a description), measured from the README read1,249 tokens for 15 tool definitions (15 with a description), measured from the docs page read
StatusExists under this name on 10 Nov 2026; v0.8.0 released 21 Oct 2026 per GitHub releases and PyPI.Exists on 10 Nov 2026; built into the OpenMetadata server (not a separate package). Tool access is governed by the caller's role and policies; no read-only switch is documented on the pages read.
Read10 Nov 202610 Nov 2026

In one sentence each

  • DataHub MCP Server: DataHub's server returns column-level lineage with paths and SQL transformations, plus glossary, domain and owner discovery, with per-user tokens in HTTP mode.
  • OpenMetadata MCP Server: OpenMetadata's built-in server offers per-user OAuth, persona-scoped context, table and optional column lineage, and write tools for lineage, entities and tests.

The read facts, topic by topic

Each cell joins the facts read for that topic; the full list with sources is on each entry. Read 10 Nov 2026.
TopicDataHub MCP ServerOpenMetadata MCP Server
Publisher and releasesmcp-server-datahub is published from the acryldata GitHub organization (DataHub's company) and is linked from DataHub's own docs at docs.datahub.com. The GitHub release page shows v0.8.0 as a published release (PyPI upload 21 Oct 2026). The package is on PyPI as mcp-server-datahub 0.8.0.The MCP server is built into OpenMetadata and installed and enabled by default. OpenMetadata releases 2.0.7 on 27 Oct 2026 (GitHub release page); the MCP server version follows the platform release. The release timestamp is 27 Oct 2026.
Clients, auth and scopingThe server is built to serve DataHub context to AI agents: search, lineage, query history and entity metadata. In the Docker HTTP deployment each MCP client supplies its own DataHub token, so DataHub permissions and audit identity are preserved per user. An optional get_me tool (TOOLS_IS_USER_ENABLED=true) returns the authenticated user's profile and group memberships.Get_user_context returns the authenticated caller's identity, roles, teams, domains and active persona, with no parameter to look up another user. A personal access token gives the agent the same role and access policy as the user. OAuth 2.0 runs tool calls with the user's existing SSO login (Google, Azure, Okta, Auth0, LDAP, SAML).
Access and safetyMutation tools are disabled by default and need TOOLS_IS_MUTATION_ENABLED=true, so the default is read-only. The HTTP entry point refuses to start if a shared server-wide token is configured, forcing per-user tokens. Tokens are accepted only in the Authorization header; query-string tokens are rejected.Every tool is labeled Read or Write in the reference. Unattended agents should use a Bot service account with a role-based access policy, kept separate from personal logins and auditable. The transport is Streamable HTTP (MCP spec 26 Mar 2025).
What it reachesThe README says the server shows data in databases, lakes, warehouses and BI tools registered in DataHub; no connector list is on this page.The reference lists asset types: database, dashboard, pipeline, topic, mlmodel, container, searchIndex, storedProcedure and API services; no named vendor list is on this page.
Lineage and impactDataHub's MCP server exposes a get_lineage tool covering datasets, columns and dashboards with hop control. The README says lineage is traceable at table and column level and that transformation logic is understood. A get_lineage_paths_between tool returns exact lineage paths between two assets or columns, including intermediate transformations and SQL query information. Get_dataset_queries fetches real SQL queries that reference a dataset or column.Get_entity_lineage returns upstream and downstream lineage for any entity, with an includeColumnLineage option (default false, table level only), up to 10 hops. Create_lineage lets an agent write lineage edges, and root_cause_analysis is a lineage-and-impact tool. None found: Documentation that the MCP lineage tools return transformation SQL or logic.
Definitions, ownership and rulesDataHub's MCP server lets agents discover domains, owners, tags and glossary terms. Optional mutation tools (off by default) add glossary terms, tags, owners, domains, descriptions and structured properties; this is metadata writing, and the docs do not say terms propagate along lineage through MCP. None found: Any statement in the DataHub MCP README that tags, terms or classifications propagate along lineage when read through the MCP tools.Search_metadata filters on owners, tags, tiers and glossary terms and entity types include glossary, glossaryTerm, metric, domain and dataProduct. Find_context and get_persona_context serve shared AI context documents curated per persona. Patch_entity and create_entity write metadata (create never modifies an existing entity; patch handles edits).
Freshness and upkeepNone found: Any freshness or ingestion-cadence statement in the DataHub MCP README (the server reads whatever DataHub has ingested). The README only describes caching of verified tokens for five minutes.None found: Freshness or collection-cadence statement on the OpenMetadata MCP pages.
ProfileDataHub describes itself as an open-source context platform; the MCP server is Apache-2.0 licensed Python.OpenMetadata's MCP server is described as embedded in an open knowledge graph of metadata, with supported SSO providers including Google, Azure, Okta, Auth0, LDAP and SAML.

Tool names, as published

DataHub MCP Server

search get_lineage get_dataset_queries get_entities get_aspect_history list_schema_fields get_lineage_paths_between add_tags remove_tags add_terms remove_terms add_owners remove_owners set_domains remove_domains update_description add_structured_properties remove_structured_properties get_me search_documents grep_documents save_document

OpenMetadata MCP Server

search_metadata semantic_search company_context get_entity_details get_user_context get_persona_context find_context get_entity_lineage create_lineage root_cause_analysis create_entity describe_entity_type patch_entity get_test_definitions create_test_case

Sources

  1. raw.githubusercontent.com/acryldata/mcp-server-datahub/main/README.md, read 10 Nov 2026
  2. pypi.org/pypi/mcp-server-datahub/json, read 10 Nov 2026
  3. github.com/acryldata/mcp-server-datahub/releases/tag/v0.8.0, read 10 Nov 2026
  4. docs.open-metadata.org/v2.0.x/how-to-guides/mcp/reference.md, read 10 Nov 2026
  5. docs.open-metadata.org/v2.0.x/how-to-guides/mcp.md, read 10 Nov 2026
  6. github.com/open-metadata/OpenMetadata/releases/tag/2.0.7-release, read 10 Nov 2026
  7. docs.open-metadata.org/v2.0.x/how-to-guides/mcp/connect.md, read 10 Nov 2026

Last reviewed 10 Nov 2026.