home / mcp / dataverse mcp server

Dataverse MCP Server

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

You run a lightweight, read-only MCP server that exposes live Microsoft Dataverse data through a simple MCP interface. It lets you ask natural language questions and get answers drawn from Dataverse without writing SQL.

How to use

Once you have the MCP server running, you use an MCP client to access the Dataverse data. The server provides a set of tools you can invoke to inspect the data source, list tables and columns, or run read-only queries. You can ask questions like what is the correlation between opportunities and account industry, or which calendar events you have today, and the client will fetch live results from Dataverse through the MCP bridge.

The available tools expose common data-access operations. Use the server name you configured as the prefix for tool calls. You can retrieve the list of tables, get column information for a table, and run a SELECT-like query against Dataverse data. The system returns results in a structured format suitable for conversational AI clients.

How to install

Prerequisites: Java Runtime Environment (JRE) and a Java build tool (Maven) are required to build and run the MCP server JAR. You will also need the CData JDBC Driver for Microsoft Dataverse installed and licensed, plus a Dataverse connection configuration file (.prp).

Step 1: Clone the project and move into the directory.

git clone https://github.com/cdatasoftware/microsoft-dataverse-mcp-server-by-cdata.git
cd microsoft-dataverse-mcp-server-by-cdata

Step 2: Build the MCP server JAR with dependencies.

mvn clean install

Step 3: Install and license the CData JDBC Driver for Microsoft Dataverse. Follow the driver’s installation process and license it using the jar in your lib folder.

Step 4: Configure the JDBC connection. Use the Connection String utility provided by the driver to test and obtain a JDBC URL, then copy the connection string for use in the MCP prop file.

Step 5: Create a .prp file (for example microsoft-dataverse.prp) with the JDBC connection details and driver information. Include the Prefix, ServerName, ServerVersion, DriverPath, DriverClass, JdbcUrl, and Tables fields.

Prefix=cds
ServerName=CDataCDS
ServerVersion=1.0
DriverPath=PATH\TO\cdata.jdbc.cds.jar
DriverClass=cdata.jdbc.cds.CDSDriver
JdbcUrl=jdbc:cds:InitiateOAuth=GETANDREFRESH;
Tables=

Run the MCP server

Start the MCP server using the compiled JAR and the .prp file you created. The server runs on stdio, which means you run it on the same machine as your MCP client.

java -jar /PATH/TO/CDataMCP-jar-with-dependencies.jar /PATH/TO/microsoft-dataverse.prp

Configure Claude Desktop (or another MCP client)

Add a local MCP server entry in your MCP client configuration. Provide the Java runtime as the command and pass the MCP JAR and the .prp file as arguments so the client can start the server on demand.

Available tools

dataverse_get_tables

Retrieves a list of accessible tables from the Dataverse data source. Output is a CSV where the first row contains column headers.

dataverse_get_columns

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

dataverse_run_query

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