SERVER, read 10 Nov 2026
DataHub MCP Server: 22 tools, remote and local, from Acryl Data
MCP servers for DataHubBack to the directory
- Data tool
- DataHub
- Publisher
- Acryl Data (DataHub)
- Kind
- Official
- Repository or docs
- github.com/acryldata/mcp-server-datahub
- Version or commit
- 0.8.0
- Last release
- 21 Oct 2026
- Transport
- Remote and local
- Auth
- Stdio: DATAHUB_GMS_URL + DATAHUB_GMS_TOKEN (or ~/.datahubenv); HTTP: per-request DataHub personal access token as Authorization Bearer, no shared server token.
- Access
- Configurable
- Tools listed
- 22 named in the evidence
- Tool definitions read at
- github.com/acryldata/mcp-server-datahub/blob/main/README.md
- Context weight
- 567 tokens for 22 tool definitions (22 with a description), measured from the README read
- Status
- Exists under this name on 10 Nov 2026; v0.8.0 released 21 Oct 2026 per GitHub releases and PyPI.
- Read
- 10 Nov 2026
In one sentence
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.
What DataHub MCP Server exposes
22 tools are named in the evidence read on 10 Nov 2026.
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
Not found in the sources read
- Any statement in the DataHub MCP README that tags, terms or classifications propagate along lineage when read through the MCP tools. Source.
- 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. Source.
The evidence on DataHub MCP Server, by topic
Publisher and releases
- mcp-server-datahub is published from the acryldata GitHub organization (DataHub's company) and is linked from DataHub's own docs at docs.datahub.com. (observed in the source; source, read 10 Nov 2026.)
- The GitHub release page shows v0.8.0 as a published release (PyPI upload 21 Oct 2026). (observed in the source; source, read 10 Nov 2026.)
- The package is on PyPI as mcp-server-datahub 0.8.0. (observed in the source; source, read 10 Nov 2026.)
Clients, auth and scoping
- The server is built to serve DataHub context to AI agents: search, lineage, query history and entity metadata. (Acryl Data's own claim; source, read 10 Nov 2026.)
- In the Docker HTTP deployment each MCP client supplies its own DataHub token, so DataHub permissions and audit identity are preserved per user. (Acryl Data's own claim; source, read 10 Nov 2026.)
- An optional get_me tool (TOOLS_IS_USER_ENABLED=true) returns the authenticated user's profile and group memberships. (Acryl Data's own claim; source, read 10 Nov 2026.)
Access and safety
- Mutation tools are disabled by default and need TOOLS_IS_MUTATION_ENABLED=true, so the default is read-only. (Acryl Data's own claim; source, read 10 Nov 2026.)
- The HTTP entry point refuses to start if a shared server-wide token is configured, forcing per-user tokens. (Acryl Data's own claim; source, read 10 Nov 2026.)
- Tokens are accepted only in the Authorization header; query-string tokens are rejected. (Acryl Data's own claim; source, read 10 Nov 2026.)
What it reaches
- The README says the server shows data in databases, lakes, warehouses and BI tools registered in DataHub; no connector list is on this page. (Acryl Data's own claim; source, read 10 Nov 2026.)
Lineage and impact
- DataHub's MCP server exposes a get_lineage tool covering datasets, columns and dashboards with hop control. (Acryl Data's own claim; source, read 10 Nov 2026.)
- The README says lineage is traceable at table and column level and that transformation logic is understood. (Acryl Data's own claim; source, read 10 Nov 2026.)
- A get_lineage_paths_between tool returns exact lineage paths between two assets or columns, including intermediate transformations and SQL query information. (Acryl Data's own claim; source, read 10 Nov 2026.)
- Get_dataset_queries fetches real SQL queries that reference a dataset or column. (Acryl Data's own claim; source, read 10 Nov 2026.)
Definitions, ownership and rules
- DataHub's MCP server lets agents discover domains, owners, tags and glossary terms. (Acryl Data's own claim; source, read 10 Nov 2026.)
- 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. (Acryl Data's own claim; source, read 10 Nov 2026.)
Profile
- DataHub describes itself as an open-source context platform; the MCP server is Apache-2.0 licensed Python. (Acryl Data's own claim; source, read 10 Nov 2026.)
Read next
Sources
- raw.githubusercontent.com/acryldata/mcp-server-datahub/main/README.md, read 10 Nov 2026
- pypi.org/pypi/mcp-server-datahub/json, read 10 Nov 2026
- github.com/acryldata/mcp-server-datahub/releases/tag/v0.8.0, read 10 Nov 2026
Last reviewed 10 Nov 2026.