home / mcp / cdata mcp server for odata

CData MCP Server for OData

Provides read-only live OData data access via MCP, enabling natural-language queries over data sources using the CData JDBC Driver for OData.

Installation
Add the following to your MCP client configuration file.

Configuration

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

You can run a local, read-only MCP server that exposes live OData data to large language models. This server wraps the CData JDBC Driver for OData so you can ask natural language questions and retrieve up-to-date data without writing SQL. It is designed for environments where you want a self-contained, on‑premise MCP endpoint that your client can query from the same machine or across a trusted network.

How to use

You will configure a local MCP server and run it via a standard Java command. The server exposes a small set of read operations that let your AI client list available tables and columns, and run read queries against your OData source. After starting the server, your client can ask questions like “What is the status of my Salesforce accounts?” and receive structured results. Use the built-in MCP tools by sending requests to the server through your MCP client, and rely on the client to present the results in natural language.

How to install

# Prerequisites
- Java runtime (JRE/JDK) installed
- Access to a machine where you will run the MCP server

# Step 1: Build the MCP server
mvn clean install

# Step 2: Run the MCP server (example)
java -jar /PATH/TO/CDataMCP-jar-with-dependencies.jar /PATH/TO/odata.prp

Additional notes

The MCP server is designed to run locally and uses stdio for communication, meaning the client and server reside on the same machine or within the same secured environment. You will typically provide a .prp configuration file that contains your JDBC driver path, driver class, and the JDBC connection string.

Available tools

odata_get_tables

Retrieves a list of tables available in the data source. The output is returned in CSV format with column headers on the first line.

odata_get_columns

Retrieves a list of columns for a specific table. The output is returned in CSV format with column headers on the first line.

odata_run_query

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