Log inSubmit server
DBHub Analytics logo

DBHub Analytics

Query and inspect Databricks, BigQuery, and six relational database systems through MCP.

InstallRepositoryWebsite
Full README Compare alternatives Embed widget Claim this listing

Install DBHub Analytics

Pick your client, then add it in one click or paste the config.

STDIONode.js
~/Library/Application Support/Claude/claude_desktop_config.json
claude_desktop_config.json
{
  "mcpServers": {
    "dbhub-analytics": {
      "command": "npx",
      "args": [
        "-y",
        "dbhub-analytics"
      ]
    }
  }
}

Settings → Developer → Edit Config, merge this into mcpServers, then fully quit and reopen Claude Desktop.

We couldn’t confirm it starts correctly

We ran the install command but it didn't respond within our test window — this can mean a slow first-time install rather than a real problem.

npx -y dbhub-analytics

No response to initialize.

This is an experimental automated check and can have false negatives — missing environment variables, a slow cold install, etc. It doesn’t necessarily mean something’s wrong. Last checked .

Overview

DBHub Analytics MCP server connects MCP-compatible AI clients to Databricks SQL, Google BigQuery, PostgreSQL, MySQL, MariaDB, SQL Server, and SQLite. It exposes SQL execution with read-only enforcement and row limits, plus metadata search for catalogs, schemas, tables, columns, indexes, and routines. Use it when an agent needs controlled database analysis, schema exploration, diagnostics, or access to multiple database sources through one configuration. Connections use database-specific DSNs or a TOML file, with stdio and HTTP startup options documented.

Use cases

  • Query Databricks or BigQuery data with an AI assistant
  • Explore database schemas and object metadata
  • Diagnose backend issues with read-only SQL
  • Inspect production read replicas with row limits
  • Compare information across multiple configured databases

Key features

  • Read-only SQL execution with row limiting
  • Catalog, schema, table, and column discovery
  • Support for seven database systems
  • Multiple database sources through TOML
  • Databricks and BigQuery connectivity
  • Optional custom parameterized SQL tools

Tools2

What DBHub Analytics lets your agent do. Expand a tool for its parameters.

Self-reported~38 tokens of context

Security & health

What it can reach and how reliably it runs — from our own checks, not the maintainer’s claims.

ActiveLast checked

Recent health check succeeded.

20/23 checks healthy over the last 95d

Supply-chain signal

No high-severity advisories surfaced by our automated scan.

Critical 0High 0Medium 0Low 0

Scanned 10d ago via OSV.dev · dbhub-analytics (npm)

Adoption & maintenance

Factual signals from GitHub, npm, and our automated checks — not a rating.

npm downloads204Package downloads in the last 30 days.
Last commit7mo agoMost recent push to the default branch.
Install checkInconclusiveDidn't respond in our test window — often a slow first install.
Tools exposed2Callable tools this server registers over MCP.
Directory activity3 viewsConfig copies, upvotes, and views on AllMCPs.

How DBHub Analytics works

What DBHub Analytics MCP server does

DBHub Analytics MCP server provides a single MCP interface for querying and inspecting several database systems. Supported connections include PostgreSQL, MySQL, MariaDB, SQL Server, SQLite, Databricks SQL warehouses, and Google BigQuery datasets. An MCP-compatible assistant can use the server to send SQL statements, inspect database structure, and work with metadata without requiring a separate MCP server for each database type.

The project is built on DBHub and is intended for analytical and operational database access through AI clients. It can be used with local databases, cloud warehouses, or multiple sources configured together. The available interface is deliberately focused: SQL execution and object discovery are the core operations.

How it works

The server starts with either a database connection string or a TOML configuration file. A DSN identifies the database type and connection details, such as the host, port, database, warehouse, project, dataset, or authentication parameters. For multiple sources, a TOML file defines each source with its own identifier and connection settings.

Clients communicate with the process through MCP. Client configuration examples are provided for Claude Desktop, Claude Code, Cursor, VS Code Copilot, and Windsurf.

SQL access includes read-only enforcement and row limiting. The project also describes query timeouts, SSH tunneling, SSL/TLS encryption, and service-account authentication as supported configuration or access features. These controls make the server suitable for querying read replicas or analytical systems where unrestricted writes and unbounded result sets should be avoided.

Setup and configuration

The simplest setup passes a DSN to the npm package:

Terminal
npx dbhub-analytics --dsn "postgres://user:password@localhost:5432/dbname"

Use a database-specific DSN for Databricks or BigQuery. BigQuery connections can specify a GCP project, dataset, service-account key file, and location. SQLite accepts a file path or an in-memory database.

For several databases, create dbhub.toml with a [[sources]] entry for each connection, then start the server with --config. This lets one MCP configuration expose Databricks, BigQuery, PostgreSQL, or other supported sources together. A demo mode is also available through npx dbhub-analytics --demo.

Tools and capabilities

DBHub Analytics MCP server exposes these MCP tools:

  • execute_sql: Runs SQL with read-only enforcement and row limiting.
  • search_objects: Searches or browses catalogs, schemas, tables, columns, indexes, and routines.

The configuration can also define reusable, parameterized SQL queries as custom tools in dbhub.toml. This is useful when an application needs a known query shape rather than unrestricted ad hoc SQL. Schema discovery follows a progressive path from broader objects such as catalogs or datasets toward tables and columns.

Limitations and notes

Read-only behavior is a stated safeguard, so this server should not be selected when an agent must insert, update, delete, or administer database resources.

Authentication and network access remain dependent on the target database and its connection settings. Examples include credentials in DSNs, BigQuery service-account files, SSL/TLS options, and SSH tunneling. Protect configuration files and connection strings because they may contain sensitive access information.

The README names stable support for seven database protocols but does not describe feature parity for every database engine. SQL syntax, metadata availability, authentication behavior, and warehouse-specific options may therefore vary by source. The DBHub Analytics MCP server is best evaluated against the particular database and client configuration you plan to use.

Read the full README View source on GitHub

Reviews

No reviews yet — be the first to share how it worked for you.

Frequently Asked Questions about DBHub Analytics

DBHub Analytics is an MCP server that connects AI clients to Databricks SQL, Google BigQuery, PostgreSQL, MySQL, MariaDB, SQL Server, and SQLite. Its main tools execute SQL with read-only and row-limit safeguards and search database objects such as schemas, tables, columns, indexes, and routines.

Install and run it through the npm package with `npx dbhub-analytics --dsn "<connection string>"`. You can also run the published Docker image, use demo mode, or start it with a TOML configuration for multiple database sources.

Yes, DBHub Analytics works with Cursor through its MCP settings. Add a server entry that runs `npx -y dbhub-analytics` and supplies a supported `--dsn`, or pass `--config` with the path to a multi-database TOML file.

DBHub Analytics supports PostgreSQL, MySQL, MariaDB, SQL Server, SQLite, Databricks SQL, and Google BigQuery. Each connection uses a corresponding DSN protocol, including `postgres://`, `mysql://`, `mariadb://`, `sqlserver://`, `sqlite:///`, `databricks://`, or `bigquery://`.

Yes, it can connect to multiple databases through one TOML configuration file. Define several `[[sources]]` entries, give each source an ID and connection settings, then launch the server with `--config dbhub.toml`.

Related MCP servers

All in Databases All alternatives
Add a badge to your READMEShow it’s listed — live SVG, Markdown or HTML.

AllMCPs Directory Badge

Full Badge Customizer

Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.

Badge Style:
Live Dynamic SVG PreviewDBHub Analytics AllMCPs Directory Badge
Markdown (GitHub README)
[![AllMCPs](https://allmcps.com/api/badge/dbhub-analytics?style=directory)](https://allmcps.com/mcp/dbhub-analytics)
HTML Embed
<a href="https://allmcps.com/mcp/dbhub-analytics"><img src="https://allmcps.com/api/badge/dbhub-analytics?style=directory" alt="DBHub Analytics on AllMCPs" /></a>