mirror of https://github.com/googleapis/genai-toolbox.git synced 2026-04-09 03:02:26 -04:00

Files

Yuan Teoh 975d02e243 docs: update docs for antigravity (#2015 )

Update docs formatting and bullet point indentation. 
Update option to move binary into executables folder.

2025-11-21 15:57:07 -08:00

4.0 KiB

Raw Blame History

Cloud Spanner MCP Server

The Cloud Spanner Model Context Protocol (MCP) Server gives AI-powered development tools the ability to work with your Google Cloud Spanner databases. It supports executing SQL queries and exploring schemas.

Features

An editor configured to use the Cloud Spanner MCP server can use its AI capabilities to help you:

Query Data - Execute DML and DQL SQL queries
Explore Schema - List tables and view schema details

Installation and Setup

Prerequisites

Download and install MCP Toolbox:
1. Download the Toolbox binary: Download the latest binary for your operating system and architecture from the storage bucket. Check the releases page for OS and CPU architecture support: https://storage.googleapis.com/genai-toolbox/v0.21.0/<os>/<arch>/toolbox
  - Replace <os> with linux, darwin (macOS), or windows.
  - Replace <arch> with amd64 (Intel) or arm64 (Apple Silicon).
```
curl -L -o toolbox https://storage.googleapis.com/genai-toolbox/v0.21.0/linux/amd64/toolbox
```
2. Make it executable:
```
chmod +x toolbox
```
3. Move binary to /usr/local/bin/ or /usr/bin/:
```
sudo mv toolbox /usr/local/bin/
# sudo mv toolbox /usr/bin/
```
  On Windows, move binary to the WindowsApps\ folder:
```
move "C:\Users\<path-to-binary>\toolbox.exe" "C:\Users\<username>\AppData\Local\Microsoft\WindowsApps\"
```
  Tip: Ensure the destination folder for your binary is included in your system's PATH environment variable. To check PATH, use echo $PATH (or echo %PATH% on Windows).
  
  Note: You may need to restart Antigravity for changes to take effect.
A Google Cloud project with the Cloud Spanner API enabled.
Ensure Application Default Credentials are available in your environment.
IAM Permissions:
- Cloud Spanner Database User (roles/spanner.databaseUser) (for data access)
- Cloud Spanner Viewer (roles/spanner.viewer) (for schema access)

Configuration

The MCP server is configured using environment variables.

export SPANNER_PROJECT="<your-gcp-project-id>"
export SPANNER_INSTANCE="<your-spanner-instance-id>"
export SPANNER_DATABASE="<your-spanner-database-id>"
export SPANNER_DIALECT="googlesql" # Optional: "googlesql" or "postgresql". Defaults to "googlesql".

Add the following configuration to your MCP client (e.g., settings.json for Gemini CLI):

{
  "mcpServers": {
    "spanner": {
      "command": "toolbox",
      "args": ["--prebuilt", "spanner", "--stdio"],
      "env": {
        "SPANNER_PROJECT": "your-project-id",
        "SPANNER_INSTANCE": "your-instance-id",
        "SPANNER_DATABASE": "your-database-name",
        "SPANNER_DIALECT": "googlesql"
      }
    }
  }
}

Usage

Once configured, the MCP server will automatically provide Cloud Spanner capabilities to your AI assistant. You can:

"Execute a DML query to update customer names."
"List all tables in the my-database."
"Execute a DQL query to select data from orders table."

Server Capabilities

The Cloud Spanner MCP server provides the following tools:

Tool Name	Description
`execute_sql`	Use this tool to execute DML SQL.
`execute_sql_dql`	Use this tool to execute DQL SQL.
`list_tables`	Lists detailed schema information for user-created tables.

Documentation

For more information, visit the Cloud Spanner documentation.

4.0 KiB Raw Blame History