home / mcp / google drive mcp server

Google Drive MCP Server

This read-only MCP Server allows you to connect to Google Drive data from Claude Desktop through CData JDBC Drivers. For full CRUD support, check out the first managed MCP platform: CData Connect AI (https://www.cdata.com/ai/).

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "cdatasoftware-google-drive-mcp-server-by-cdata": {
      "command": "java",
      "args": [
        "-jar",
        "/PATH/TO/CDataMCP-jar-with-dependencies.jar",
        "/PATH/TO/google-drive.prp"
      ]
    }
  }
}

You can run a local Google Drive MCP server that exposes Google Drive data as read-only MCP tools. This lets you ask natural-language questions and receive live results from Google Drive without writing SQL. The server is started locally and accessed by your MCP client on the same machine.

How to use

You will connect your MCP client to the local server to read data from Google Drive. Start the server on the same machine as your client, then configure your client to load the new MCP server definition. Once configured, you can ask questions like “What Google Drive files are shared with user X?” or “Show me the latest modified documents in folder Y.” The server presents a set of tools that allow you to list available tables and columns and to run queries, all through natural language questions processed by your AI client.

Typical usage flow: - Start the MCP server on your machine. - Ensure your MCP client (for example, Claude Desktop or another compatible client) has the new server configured. - Ask the AI client questions about your Google Drive data. The client will use the built-in tools to read tables, inspect columns, and retrieve results from Google Drive without exposing raw SQL to you.

How to install

Prerequisites: - Java (JDK) installed on your machine - A Java-enabled MCP client setup (Claude Desktop or similar) - Optionally Maven if you want to build locally from source - Access to the Google Drive API via the CData JDBC Driver for Google Drive (license required)

Step-by-step commands and actions you will perform: 1) Build the MCP server from source (if you are starting from local code): ``# Clone the project and build the MCP server git clone https://github.com/cdatasoftware/google-drive-mcp-server-by-cdata.git cd google-drive-mcp-server-by-cdata mvn clean install`` This creates the MCP server JAR named CDataMCP-jar-with-dependencies.jar.

2) Install the CData JDBC Driver for Google Drive following the driver installer procedures and license it when prompted. This enables Google Drive as a relational data source.

3) Create a JDBC connection configuration (example shown). Save the resulting connection string for later use in the server configuration file (.prp). You will typically run the JDBC driver’s connection utility to generate this string, then copy it into your .prp file.

4) Create the MCP server properties file (google-drive.prp) with the required fields, including the JDBC connection string you copied. Example fields include Prefix, ServerName, ServerVersion, DriverPath, DriverClass, JdbcUrl, and Tables. You will keep Tables blank to access all data or specify particular tables.

5) Run the MCP server using the final start command (below). The server runs in stdio mode and must be started on the same machine as your MCP client.

6) Configure your MCP client to load the new server. Create or update the client’s MCP config to include the new server entry, pointing to the jar and the .prp file you created.

Additional configuration and notes

The server exposes Google Drive data through a simple MCP interface. You don’t need to call the individual tools directly; simply pose questions to your AI client and rely on the built-in tools to retrieve tables, columns, and query results.

Example tool endpoints you will see available after configuration include:
- {servername}_get_tables: Retrieves a list of tables available in the data source. Output is CSV with column headers on the first line.
- {servername}_get_columns: Retrieves a list of columns for a table. Output is CSV with column headers on the first line.
- {servername}_run_query: Executes a SQL SELECT query against the data source.

Running the server

java -jar /PATH/TO/CDataMCP-jar-with-dependencies.jar /PATH/TO/Salesforce.prp

Troubleshooting

If you cannot see your MCP server in your client, ensure you have fully exited and restarted the client so the new server appears. Verify that the connection string in the .prp file is correct and test the connection from the driver utility. If issues persist, confirm that the Google Drive data source is accessible and that OAuth authentication (if used) completes in your browser.

License and support

This MCP server is distributed under the MIT License. You are free to use, modify, and distribute the software in compliance with the license terms. For help and community discussions, reach out to the CData Community.

Available tools

googledrv_get_tables

Retrieves a list of tables available in the Google Drive data source. Output is in CSV format with column headers.

googledrv_get_columns

Retrieves a list of columns for a specified table. Output is in CSV format with column headers.

googledrv_run_query

Executes a SQL SELECT query against the Google Drive data source.