home / mcp / google sheets mcp server

Google Sheets MCP Server

This read-only MCP Server allows you to connect to Google Sheets 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-sheets-mcp-server-by-cdata": {
      "command": "java",
      "args": [
        "-jar",
        "/PATH/TO/CDataMCP-jar-with-dependencies.jar",
        "/PATH/TO/google-sheets.prp"
      ]
    }
  }
}

You can run a local, read-only MCP Server that exposes Google Sheets data through a simple MCP interface. This lets you ask natural language questions about your live Sheets data and get direct results without writing SQL. The server leverages the CData JDBC Driver to treat Sheets as relational models and serves them via MCP so client tools can query them easily.

How to use

You will interact with the Google Sheets MCP Server through an MCP client. Start the local MCP server, then connect your client to the running process. Once connected, you can ask questions like: - What is the current open status by sheet or table? - How many rows match a given condition across multiple sheets? - Which accounts or records meet specific criteria today? The server exposes tools that let you list available tables, list columns, and run read-only queries. You do not need to know SQL to use these capabilities; you can rely on natural language prompts and the MCP client will translate your questions into queries against Google Sheets data.

How to install

Prerequisites: you need a Java Runtime Environment (JRE) or JDK installed, and you will use Maven to build the server.

1) Clone the project and enter the directory.

git clone https://github.com/cdatasoftware/google-sheets-mcp-server-by-cdata.git
cd google-sheets-mcp-server-by-cdata

2) Build the server to produce the MCP JAR.

mvn clean install

3) Install and license the CData JDBC Driver for Google Sheets. You will download the driver from the vendor and license it locally using the provided jar.

4) Create a JDBC connection property file (google-sheets.prp) with your connection details. The file should include your driver path, driver class, and the JDBC URL you obtain from the connection utility.

5) Run the MCP server with the prepared property file.

java -jar /PATH/TO/CDataMCP-jar-with-dependencies.jar /PATH/TO/google-sheets.prp

Additional setup and notes

This server runs locally in stdio mode, which means it is intended to be used with clients running on the same machine. It exposes a minimal, read-only set of capabilities suitable for querying live Google Sheets data.

Troubleshooting

If you cannot see or start the MCP server, ensure the Java runtime is correctly installed and that you are using the exact command shown above to run the JAR with the path to your prp file.

If the client cannot retrieve data, double-check that the JDBC connection string was created successfully and that the produced google-sheets.prp is correctly referenced by the MCP start command.

If you need help, contact the CData Support Team for driver or server questions.

Notes

This setup targets a local, read-only MCP Server for Google Sheets using the CData JDBC Driver. It is designed for environments where you want live data access from Sheets without SQL and without remote MCP hosting.

Available tools

google_sheets_get_tables

Retrieves a list of tables available in the data source. Output is CSV with headers on the first line.

google_sheets_get_columns

Retrieves the list of columns for a specified table. Output is CSV with headers on the first line.

google_sheets_run_query

Executes a SQL SELECT query against the data source and returns results.