Skip to content
MCP server

MCP Toolbox for Databases

Open source database MCP server with prebuilt tools and a framework for custom ones

Pricing
Open source
Platforms
CLI, self-hosted, API
Runs
Remotely
Version
1.17.0
Listing updated
2 days ago

About MCP Toolbox for Databases

MCP Toolbox for Databases is an open-source MCP server for databases. It runs as a hosted remote server, so there is nothing to install locally.

The project started life as the Gen AI Toolbox for Databases and was renamed when it aligned with MCP. It does two jobs. Run with a --prebuilt=<database> flag and it gives an MCP client generic tools such as list_tables and execute_sql for a long list of engines, among them AlloyDB, BigQuery, Cloud SQL, Spanner, Firestore, PostgreSQL, MySQL, SQL Server, Oracle, MongoDB, Redis, Elasticsearch, ClickHouse, Neo4j and Snowflake. A toolset suffix such as postgres/data loads only part of the tool group.

The second job is a custom tools framework. A tools.yaml file declares sources, tools with typed parameters and fixed SQL statements, toolsets and prompts, so production agents see only structured, parameterised actions instead of free-form SQL. The README says it handles connection pooling, IAM-based authentication and OpenTelemetry metrics and tracing.

SDKs for Python, JavaScript, Go and Java let applications built on LangChain, LlamaIndex or Google's Agent Development Kit load the same tools. The registry lists an OCI image served over streamable HTTP, and the README notes a managed Google Cloud MCP option for those who prefer it.

Install

Run in a terminal

claude mcp add mcp-toolbox -- docker run -i --rm us-central1-docker.pkg.dev/database-toolbox/toolbox/toolbox:1.17.0

Add to claude_desktop_config.json

{
  "mcpServers": {
    "mcp-toolbox": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "us-central1-docker.pkg.dev/database-toolbox/toolbox/toolbox:1.17.0"
      ]
    }
  }
}
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "mcp-toolbox": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "us-central1-docker.pkg.dev/database-toolbox/toolbox/toolbox:1.17.0"
      ]
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "mcp-toolbox": {
      "type": "stdio",
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "us-central1-docker.pkg.dev/database-toolbox/toolbox/toolbox:1.17.0"
      ]
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "mcp-toolbox": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "us-central1-docker.pkg.dev/database-toolbox/toolbox/toolbox:1.17.0"
      ]
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "mcp-toolbox": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "us-central1-docker.pkg.dev/database-toolbox/toolbox/toolbox:1.17.0"
      ]
    }
  }
}

Tools this server provides (2)

Taken from the project documentation; the running server may differ.

list_tables
List tables in the connected database
execute_sql
Run a SQL statement against the connected database

What it does

Prebuilt database tools
Start with one flag to get schema exploration and SQL execution for a chosen engine.
Broad engine support
Covers Google Cloud databases plus PostgreSQL, MySQL, Oracle, MongoDB, Redis, Snowflake and more.
Custom tools in YAML
Declare sources, parameterised statements, toolsets and prompts in tools.yaml.
Pooling and IAM auth
Connection pooling and integrated authentication are handled by the server.
OpenTelemetry built in
Metrics and tracing come out of the box.
Client SDKs
Python, JavaScript, Go and Java SDKs load tools into agent frameworks.

Good for

  • Explore a database schema and run queries from an IDE assistant
  • Expose a vetted set of parameterised queries to a production agent
  • Connect an agent to BigQuery, AlloyDB or Spanner
  • Add database tools to a LangChain or LlamaIndex application
  • Trace and monitor agent database calls with OpenTelemetry

Pricing

Open source

Free under Apache 2.0

MCP Toolbox for Databases FAQ

Which databases are supported?

The README lists AlloyDB, BigQuery, Cloud SQL, Spanner, Firestore, PostgreSQL, MySQL, MariaDB, SQL Server, Oracle, MongoDB, Redis, Elasticsearch, CockroachDB, ClickHouse, Couchbase, Neo4j, Snowflake, Trino and more.

How do I restrict what an agent can run?

Define custom tools in tools.yaml with fixed statements and typed parameters instead of using the generic SQL tool.

What licence is it under?

The README badge shows Apache 2.0.

Was it called something else before?

Yes. It was originally Gen AI Toolbox for Databases, at the genai-toolbox repository.

Is MCP Toolbox for Databases free?

Yes. MCP Toolbox for Databases is open source, so you can use and self-host it for free.

How do I add MCP Toolbox for Databases to Claude Code?

Run this command in your terminal: claude mcp add mcp-toolbox -- docker run -i --rm us-central1-docker.pkg.dev/database-toolbox/toolbox/toolbox:1.17.0

What are the best alternatives to MCP Toolbox for Databases?

The closest alternatives to MCP Toolbox for Databases are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.

Alternatives to MCP Toolbox for Databases

  1. REA

    MCP server

    CLI and MCP server that lets coding agents reverse engineer apps and binaries

    Runs locally
    • Open source
    • 464K GitHub stars
    • Developer tools
  2. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  3. GitHub

    MCP server

    Official MCP server connecting AI assistants to GitHub repos, issues and workflows

    Not checked yet
    • Remote
    • Open source
    • 34.5K GitHub stars
    • Developer tools
  4. MCP toolkit giving coding agents symbol-level code retrieval and editing

    Runs locally
    • Open source
    • 31.5K GitHub stars
    • Developer tools
  5. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory
  6. Ouroboros

    MCP server

    Spec-first layer that pins requirements and verifies results of AI coding agent work

    Runs locally
    • Open source
    • 6.4K GitHub stars
    • Developer tools