mcp-server-datahub

Name: mcp-server-datahub
Author: acryldata

@acryldata

Repository Homepage

About mcp-server-datahub

The official Model Context Protocol (MCP) server for DataHub (https://datahub.com)

#datahub #metadata #ai-agents

Basic information

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "datahub": {
      "command": "uvx",
      "args": [
        "mcp-server-datahub"
      ],
      "env": {
        "DATAHUB_GMS_URL": "<your-datahub-url>",
        "DATAHUB_GMS_TOKEN": "<your-datahub-token>"
      }
    }
  }
}

Tools

No tools detected

We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.

Overview

What is mcp-server-datahub?

An MCP (Model Context Protocol) server implementation for DataHub, an open-source context platform that gives organizations a single pane of glass across their entire data supply chain, unifying data discovery, governance, and observability. It enables AI agents to search, explore lineage, understand business context, and generate SQL using natural language.

How to use mcp-server-datahub?

Configure via environment variables (e.g., TOOLS_IS_MUTATION_ENABLED, TOOLS_IS_USER_ENABLED) and invoke the provided tools: search, get_lineage, get_dataset_queries, get_entities, list_schema_fields, get_lineage_paths_between, plus mutation tools when enabled. See the DataHub MCP server docs for detailed usage instructions.

Key features of mcp-server-datahub

Structured search with wildcard, field, and boolean syntax
SQL intelligence: fetch popular queries and generate accurate SQL
Table and column-level lineage with upstream/downstream exploration
Mutation tools: add/remove tags, terms, owners, domains, descriptions, structured properties
Document tools: search, grep, save knowledge articles and FAQs
User tools: retrieve authenticated user information

Use cases of mcp-server-datahub

Find trustworthy data using natural language search across tables, columns, dashboards, and metrics
Explore data lineage to understand impact of changes before they affect downstream users
Navigate business glossaries, domains, data products, and asset relationships
Generate accurate SQL queries using context like documentation, lineage, and popular queries
Manage metadata: assign tags, glossary terms, owners, domains, and descriptions

FAQ from mcp-server-datahub

What tools does mcp-server-datahub provide?

It provides search, get_lineage, get_dataset_queries, get_entities, list_schema_fields, get_lineage_paths_between, plus mutation tools (add/remove tags, terms, owners, domains, descriptions, structured properties) and optional user and document tools.

How do I enable mutation tools?

Set the environment variable TOOLS_IS_MUTATION_ENABLED=true. By default mutation tools are disabled.

What environment variables are available for configuration?

Key variables include TOOLS_IS_MUTATION_ENABLED, TOOLS_IS_USER_ENABLED, DATAHUB_MCP_DOCUMENT_TOOLS_DISABLED, SAVE_DOCUMENT_TOOL_ENABLED, SAVE_DOCUMENT_PARENT_TITLE, TOOL_RESPONSE_TOKEN_LIMIT, ENTITY_SCHEMA_TOKEN_BUDGET, and SEMANTIC_SEARCH_ENABLED, among others.

Can I search within document content?

Yes, the grep_documents tool allows searching within document content using regex patterns. Document tools are automatically hidden if no documents exist in the catalog.

What is the maximum token limit for tool responses?

The default TOOL_RESPONSE_TOKEN_LIMIT is 80,000 tokens, and the schema token budget per entity is 16,000 tokens (ENTITY_SCHEMA_TOKEN_BUDGET).

Comments

More Other MCP servers

🚀 Model Context Protocol (MCP) Curriculum for Beginners

microsoft

This open-source curriculum introduces the fundamentals of Model Context Protocol (MCP) through real-world, cross-language examples in .NET, Java, TypeScript, JavaScript, Rust and Python. Designed for developers, it focuses on practical techniques for building modular, scalable,

16.7K