Skill Detail

Query operational databases from MCP clients with DBHub

Use DBHub to expose guarded, token-efficient database inspection and SQL tools to MCP clients across Postgres, MySQL, SQL Server, MariaDB, and SQLite.

Data Extraction & TransformationMCP
Data Extraction & Transformation MCP Security Reviewed
⭐ 3.1k GitHub stars ⬇ 78.8k/wk npm
INSTALL WITH ANY AGENT
npx skills add agentskillexchange/skills --skill query-operational-databases-from-mcp-clients-with-dbhub Copy
Works best when you want a reusable capability, not another fragile one-off prompt.
At a glance
Tools required
Node.js 22.5 or newer or Docker, database DSN/configuration, MCP-compatible client
Install & setup
Run with npx @bytebase/dbhub@latest –transport http –port 8080 –dsn or use the documented Docker image; configure read-only mode, row limits, timeouts, SSH/SSL, and custom tools as appropriate before exposing it to an MCP client.
Author
Bytebase
Publisher
Open Source Vendor
Last updated
Jul 5, 2026
Quick brief

DBHub is a zero-dependency MCP server for connecting agents to operational databases through a small, controlled tool surface. Use this skill when an MCP-compatible assistant needs to search schema objects, run bounded SQL, use reusable configured database tools, and inspect results without hand-rolling database glue for every agent.

How it works

What this skill actually does

Invoke it instead of direct ad hoc SQL access when the workflow needs MCP integration, multi-database support, read-only or timeout guardrails, and a token-efficient schema exploration path. The scope boundary is guarded database access for MCP clients; it is not a generic database client, backend platform, or SQL library listing.

Inputs and prerequisites: Node.js 22.5 or newer or Docker, database DSN/configuration, MCP-compatible client.

Setup notes: Run with npx @bytebase/dbhub@latest –transport http –port 8080 –dsn or use the documented Docker image; configure read-only mode, row limits, timeouts, SSH/SSL, and custom tools as appropriate before exposing it to an MCP client.

Source and verification boundary: use https://dbhub.ai as the canonical reference before running the workflow; keep commands, API calls, CLI usage, and generated outputs reviewable against that upstream source.

Framework fit: publish this as a MCP workflow only when the operator can invoke the documented toolchain directly, rather than treating the upstream project as a generic product listing.