Server profile

DBHub MCP Server

DBHub exposes SQL execution and database discovery through a small MCP interface. It spans five database engines, supports multiple configured connections, and includes a web workbench for use without an assistant.

Server facts

DBHub MCP Server

Project

Author
bytebase
Website
dbhub.ai
License
MIT
Category
Databases

Tech

Runtime / language
node
Transports
stdio
Remote available
No
Authentication
Not specified

Activity

GitHub stars
3,664
Last commit
Data as of

Tags

  • Database Workbench
  • Multi Connection
  • Query Controls
  • Schema Discovery
  • Sql

Overview

DBHub can be used from an assistant or directly through its browser workbench. The workbench runs database queries and configured custom operations, and lets users inspect request traces without connecting an MCP client. On the MCP side, the built-in interface concentrates work into two tools: SQL execution with transaction handling, and object discovery that reveals detail progressively. The latter covers schemas and the objects within them, including procedures, indexes, columns, and tables. Parameterized operations can also be defined as reusable custom tools in dbhub.toml. A single instance can hold multiple connections through TOML configuration, including connections to different database engines.

The supported engines are SQLite and SQL Server alongside MariaDB, PostgreSQL, and MySQL. Execution controls include query deadlines, limits on returned rows, and a read-only setting; encrypted connections and SSH tunnels are also supported. Deployment security is a separate concern from query safeguards. HTTP clients receive no authentication from DBHub itself, and its default listener binds across all interfaces. The README therefore advises restricting the listener to loopback and using a firewall or reverse proxy for production access. Host-header checks help resist DNS rebinding, with explicitly allowed names available for custom domains. Startup examples cover Docker, npm, and demo operation. The npm path needs at least Node.js 22.5.0 because DBHub relies on the runtime's SQLite module.

Install & configure

Requirements

  • npm launch: Node.js version 22.5.0 or above
  • MCP client for tool calls, or a browser for workbench interaction
Documented launch / install command
npx @bytebase/dbhub@latest --transport http --host 127.0.0.1 --port 8080 --demo

The arguments exactly follow the loopback demo example. Actual database connections use the documented --dsn example or TOML setup. Demo mode is not a configured connection to your database. Before exposing real data over HTTP, account for the lack of built-in client authentication.

Config file · macOS: ~/Library/Application Support/Claude/claude_desktop_config.json

Windows: %APPDATA%\Claude\claude_desktop_config.json

Use Settings → Developer → Edit Config when available. Do not assume a Linux configuration path; check the client documentation for your platform. Remote servers use the app’s connector flow when supported, not the local stdio file shown here. Merge this example with existing settings; do not overwrite unrelated servers.

No install command stated in the README — see repo. Check the client documentation for a supported connection method.

Read the Claude Desktop setup guide

Config file · Project: .cursor/mcp.json

User: ~/.cursor/mcp.json

The project path is relative to the workspace root. The tilde denotes your user home directory. Use Cursor’s MCP settings to confirm which configuration and scope your installed version is reading. Merge this example with existing settings; do not overwrite unrelated servers.

No install command stated in the README — see repo. Check the client documentation for a supported connection method.

Read the Cursor setup guide

Config file · Workspace: .vscode/mcp.json

For user scope, use the “MCP: Open User Configuration” command if available. Do not guess a user configuration path across operating systems, profiles, containers, or remote workspaces; check the client documentation. Merge this example with existing settings; do not overwrite unrelated servers.

No install command stated in the README — see repo. Check the client documentation for a supported connection method.

Read the VS Code setup guide

Config file · macOS/Linux: ~/.codeium/windsurf/mcp_config.json

Windows: %USERPROFILE%\.codeium\windsurf\mcp_config.json

Open the MCP configuration through Windsurf’s settings when possible to confirm the active file. Paths can differ in managed or future releases; check the client documentation if this location is not present. Merge this example with existing settings; do not overwrite unrelated servers.

No install command stated in the README — see repo. Check the client documentation for a supported connection method.

Read the Windsurf setup guide

Config file · Project scope: .mcp.json at the repository root

The project file uses mcpServers. For user scope, use claude mcp add --scope user rather than assuming .mcp.json in your home directory is the user configuration. Check the client documentation for managed settings and storage details. Merge this example with existing settings; do not overwrite unrelated servers.

No install command stated in the README — see repo. Check the client documentation for a supported connection method.

Read the Claude Code setup guide

Examples are translated from the published configuration. Supply your own credentials and verify requirements in the upstream README. Remote connectors are shown only where a supported configuration format is known.

Tools

Source: repository README. Tool descriptions are a documented snapshot, not a live capability check.

ToolWhat it does
search_objectsExplore database structure, revealing object details progressively rather than returning everything immediately.
execute_sqlSubmit SQL with transaction support and the server's execution controls.

Names, permissions, and defaults can change. Check the current upstream documentation before granting access.

Use cases

  • Inspect database objects incrementally before asking an assistant to write SQL.
  • Configure reusable parameterized queries instead of repeating SQL for routine operations.
  • Keep several database connections behind one TOML-configured instance.
  • Review query behavior and request traces through the web workbench.

Strengths

  • Separates SQL execution from discovery using only two built-in tool identifiers.
  • Includes an interactive workbench that does not depend on an MCP host.
  • Combines connection options with configurable restrictions on query execution.

Limitations

  • Network clients are not authenticated by the HTTP server.
  • The default HTTP listener is broader than loopback; exposure needs deliberate configuration.
  • npm execution depends on a relatively recent Node.js runtime, starting at 22.5.0.

Frequently asked questions

No. Its browser interface can run queries, invoke custom operations, and show traces independently of an MCP client.

DBHub reads TOML configuration for simultaneous database connections, rather than requiring a separate instance for each connection.

No. Read-only operation, row caps, and query timeouts constrain database work, but DBHub does not authenticate incoming HTTP clients.

Host validation rejects unrecognized names as a DNS-rebinding safeguard. The README provides --allowed-hosts for an intended custom name or proxy hostname.

Independent listing, not an affiliation or security review. Grant the least access needed and verify the upstream instructions before running code.

Related MCP servers

MongoDB MCP Server

Databases

MongoDB MCP Server connects AI clients to MongoDB databases and Atlas. Its tools cover queries, schema inspection, database changes, and MongoDB knowledge searches, with read-only launch examples for safer exploration.

MySQL MCP Server

Databases

Expose MySQL schema inspection, data samples, and SQL execution to MCP assistants. It supports multi-database access, optional SSH tunnels, and local or network deployment, with database permissions controlling available operations.

Neon MCP Server

Databases

Manage Neon Postgres projects, branches, queries, and migrations through an AI assistant. The hosted MCP service supports OAuth or API keys, read-only access, and project-level scoping.

MCP Toolbox for Databases

Databases

MCP Toolbox for Databases connects agents and IDEs to enterprise data through prebuilt tools or custom query definitions. It combines database access with connection pooling, authentication integration, and observability.

MCP Server for MySQL

Databases

MCP Server for MySQL lets Claude and other AI clients inspect schemas and execute SQL. It defaults to read-only access, with optional write flags, SSH tunneling, multi-database support, and result redaction.

Supabase MCP Server

Databases

Connect AI assistants to Supabase projects for SQL, schema changes, debugging, and Edge Functions. Hosted HTTP setup supports project scoping, read-only access, and selectable feature groups.

Postgres MCP Pro

Databases

Give assistants PostgreSQL schema access, SQL execution, health checks, and performance analysis. Postgres MCP Pro adds query plans and hypothetical index evaluation with configurable read-only operation.

Query MCP for Supabase

Databases

Query MCP connects coding assistants to Supabase SQL, schema changes, project management, and user administration. It includes SQL risk controls, but its author has stopped active maintenance.