home / mcp / tally mcp server

Tally MCP Server

This read-only MCP Server allows you to connect to Tally data from Claude Desktop through CData JDBC Drivers. For full CRUD support, check out our MCP Server for Tally (https://www.cdata.com/drivers/tally/download/mcp).

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "cdatasoftware-tally-mcp-server-by-cdata": {
      "command": "java",
      "args": [
        "-jar",
        "/PATH/TO/CDataMCP-jar-with-dependencies.jar",
        "/PATH/TO/Salesforce.prp"
      ],
      "env": {
        "PATH": "/usr/local/bin"
      }
    }
  }
}

You can run a read-only MCP Server that exposes live Tally data to language models via a simple MCP interface. This lets you ask natural language questions and retrieve up-to-date data from Tally without writing SQL. The server wraps the CData JDBC Driver to present Tally as relational models you can query through MCP tools.

How to use

With the MCP server running on your machine, you use an MCP-capable client to ask questions about your Tally data. The server exposes a small set of built-in tools that let the AI discover what tables exist, inspect columns, and run read-only queries. You can ask questions like which accounts exist, which fields are available for a table, or pull specific rows that match criteria. You do not need to know SQL to get meaningful live data from Tally.

The available tools are named after the data source and action, and you reference them by their tool names in your MCP client. The common workflow is to first list tables, then view columns for a chosen table, and finally execute a safe SELECT to fetch data. All responses come from the live Tally data via the JDBC bridge.

How to install

Prerequisites: Java is required to run the MCP server. You should also have Maven installed to build the project.

1) Clone the project and navigate into it.

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

2) Build the server to produce the executable JAR with all dependencies.

mvn clean install

3) Install the CData JDBC Driver for Tally and license it following the driver’s instructions, then configure a connection string for Tally. The setup steps include running the driver’s connection utility, testing the connection, and saving the JDBC URL for later use.

4) Create a properties file describing your MCP server connection. Place the following in a file named tally.prp (adjust paths to your environment).

Prefix=tally
ServerName=CDataTally
ServerVersion=1.0
DriverPath=PATH\\TO\\cdata.jdbc.tally.jar
DriverClass=cdata.jdbc.tally.TallyDriver
JdbcUrl=jdbc:tally:InitiateOAuth=GETANDREFRESH;
Tables=

5) Start the MCP server using the Java runtime and pass the path to your prp file. The example uses Salesforce.prp as the target PRP in this run command.

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

Additional notes

The server runs in stdio mode, which means it is intended to be used by clients running on the same machine. If you need to connect from a different host, you would need a different MCP configuration that explicitly supports network access, which is not included in this setup.

If you want to refresh client tooling, restart your MCP client after starting or stopping the server to ensure the MCP server is visible to the client.

Available tools

tally_get_tables

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

tally_get_columns

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

tally_run_query

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