home / mcp / salesforce pardot mcp server
Provides a read-only MCP server that exposes Salesforce Pardot data for querying via MCP without SQL.
Configuration
View docs{
"mcpServers": {
"cdatasoftware-salesforce-pardot-mcp-server-by-cdata": {
"command": "java",
"args": [
"-jar",
"PATH/TO/CDataMCP-jar-with-dependencies.jar",
"PATH/TO/salesforce-pardot.prp"
]
}
}
}This MCP Server provides a read-only interface to Salesforce Pardot data via the CData JDBC Driver, exposing live Pardot data through a simple MCP protocol so you can ask natural language questions and receive instant answers without writing SQL.
Once you have the MCP server running on your machine, you can connect with an MCP client to query Pardot data in real time. You don’t need to write SQL; simply ask questions like which opportunities are open, which accounts are associated with a given campaign, or what calendar events are scheduled for today. The server presents data tables and columns as standard MCP tools, allowing you to discover data and run reads through natural language prompts.
Key tools you’ll use include listing available tables, listing columns for a table, and executing read queries. For example, you can retrieve a list of available tables, inspect the columns of a specific table, and run a SELECT query to fetch data. These tools are surfaced with names that combine your server’s identifier and the action, making it straightforward to invoke from your MCP client.
# Prerequisites
- Java Development Kit (JDK) installed
- Maven installed
# Optional: git for cloning
# 1) Clone the MCP server repository
git clone https://github.com/cdatasoftware/salesforce-pardot-mcp-server-by-cdata.git
cd salesforce-pardot-mcp-server-by-cdata
# 2) Build the server
mvn clean install
# This creates the JAR: CDataMCP-jar-with-dependencies.jar
# 3) Install the CData JDBC Driver for Salesforce Pardot
# Visit and download from:
https://www.cdata.com/drivers/pardot/download/jdbc
# 4) License the CData JDBC Driver
# Replace PATH with your installation path
cd /path/to/driver/lib
java -jar cdata.jdbc.salesforcepardot.jar --license
# Enter your name, email, and a license key or TRIAL
# 5) Configure the data source connection
# Run the connection string utility
java -jar cdata.jdbc.salesforcepardot.jar
# Build and test the connection, then copy the connection string when connected
# 6) Create a .prp file for the JDBC connection
# Example (save as salesforce-pardot.prp)
Prefix=salesforcepardot
ServerName=CDataSalesforcePardot
ServerVersion=1.0
DriverPath=PATH\TO\cdata.jdbc.salesforcepardot.jar
DriverClass=cdata.jdbc.salesforcepardot.SalesforcePardotDriver
JdbcUrl=jdbc:salesforcepardot:InitiateOAuth=GETANDREFRESH;
Tables=
# 7) Run the MCP server locally
java -jar /PATH/TO/CDataMCP-jar-with-dependencies.jar /PATH/TO/salesforce-pardot.prpThe MCP server runs on stdio, meaning it accepts input from clients that execute on the same machine. You will start the server with the Java command shown above, passing the path to your .prp file that contains the JDBC connection settings. The server will expose Pardot data as MCP tools for querying.
If you cannot see the MCP server in your client, ensure you have fully exited any previous client, then restart the client and re-open. Verify your connection string is correct by running the JDBC Connection String utility and testing the connection. If issues persist, contact CData Support or join the CData Community for guidance.
The server exposes standard MCP tools for reading data from Pardot. You will interact with tools that retrieve table lists, column lists, and execute SQL-like queries against those tables. Each tool is named to reflect its action and server identity, for example, your server name combined with _get_tables, _get_columns, or _run_query.
If you script requests directly to the MCP server, you can use the provided tool names to invoke the corresponding actions. For example, request a list of tables, request columns for a table, or run a SELECT query against a table. Replace the server-specific prefix with your actual server name in the request payload.
License the JDBC driver before usage and keep your license key secure. Use the connection utility to ensure OAuth or other authentication flows are correctly configured. If you are integrating with a broader system, follow your security policy for credential storage and access controls.
java -jar /PATH/TO/CDataMCP-jar-with-dependencies.jar /PATH/TO/salesforce-pardot.prpRetrieves the list of available tables from the Pardot data source; output is in CSV, with the first line containing column headers.
Retrieves the list of columns for a specific table; output is in CSV, with the first line containing column headers.
Executes a SQL SELECT query against the configured Pardot data source.