# bytebase/dbhub

Token conscious database MCP server for Postgres, MySQL, SQL Server, MariaDB, SQLite.

Repository: https://github.com/bytebase/dbhub
Canonical: https://ross.abutalabs.com/products/dbhub
Homepage: https://dbhub.ai
Language: TypeScript
License: MIT
License Family: permissive
Topics: ai, anthropic, claude, database, mcp, mcp-server, claude-ai, mysql, postgres, sqlserver, cursor, mariadb, dify, sqlite, codex, llm, mssql, sql, postgresql, agents
Last push: 2026-08-21T05:20:06+00:00

## Health v2 (maintenance only)
Score: 86/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 98, release rhythm 98, longevity 38
- inputs: {"age_days": 542, "days_push": 12, "days_rel": 12, "gap_med": 5.0, "n_releases_24m": 7}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3407, forks 291 (observed 2026-08-28T04:08:03.111595+00:00)

## What it is
DBHub is a minimal, token-efficient MCP server that lets AI clients like Claude, Cursor, and VS Code connect to and explore PostgreSQL, MySQL, MariaDB, SQL Server, and SQLite databases. It exposes just two tools by default, with guardrails such as read-only mode, row limits, and query timeouts.

## Use cases
- connect claude to a postgres database via mcp
- explore database schema with cursor or vs code
- let non-technical staff query a database read-only through claude desktop
- replace separate mcp servers for each database with one process
- debug sql queries with an ai coding assistant
- run read-only production diagnostics with query guardrails

## When to choose
- you want an MCP database gateway with minimal context-window token usage
- you need one server supporting multiple database engines
- you need safety guardrails like read-only mode and query timeouts for AI database access

## When to avoid
- you need a full database administration or migration platform
- you use databases outside PostgreSQL, MySQL, MariaDB, SQL Server, or SQLite
- you need deep IDE-integrated database tooling rather than an AI client bridge

## Facets
- artifact type: service
- maturity: active
- function: mcp, database, database-driver, http-server, developer-tools
- domain: databases, large-language-models, developer-tools
- platform: cli, cross-platform
- tags: mcp-server, token-efficient, sql, claude, cursor, read-only-mode, multi-database, ai-agents, nodejs, docker

## Member repositories
- bytebase/dbhub (main) score 86

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:03.111595+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:38:33.486646+00:00, confidence not recorded.
  - readme: https://github.com/bytebase/dbhub (fetched 2026-08-28T04:08:03.111595+00:00, sha c3a90dc040f4)
  - homepage: https://dbhub.ai (fetched 2026-08-29T09:32:42.448583+00:00, sha c6b4343b7bcf)
  - site_page: https://dbhub.ai/installation (fetched 2026-08-29T09:32:42.457917+00:00, sha 65cb7881e3a3)
  - site_page: https://dbhub.ai/quickstart (fetched 2026-08-29T09:32:42.460172+00:00, sha 2b84720163b3)
- Data as of 2026-08-30T08:39:29.467469+00:00.