Salesforce Einstein by CData

Salesforce Einstein by CData is an MCP server that exposes Salesforce Einstein data as relational SQL models for AI assistants. It connects Large Language Models to live analytics and CRM insights using the CData JDBC Driver for Salesforce Einstein. Designed for data analysts, sales operations teams, and enterprise developers, this server enables users to explore Salesforce Einstein datasets through conversational prompts rather than manually constructing SQL queries. The integration operates locally over standard input and output (stdio), translating model queries into driver operations. By abstracting schema exploration and read operations into standard tools, the server lets clients discover database tables, inspect column definitions, and execute SELECT statements directly. Users can query customer insights, track metrics, and inspect account histories safely within a read-only environment that prevents accidental modifications to live enterprise data stores.

Category: CRM, ERP & E-commerce

Tags: crm, data-query, einstein, jdbc, salesforce

Visit Salesforce Einstein by CData

How to install and configure Salesforce Einstein by CData

1. Clone the repository and build the project: bash git clone https://github.com/cdatasoftware/salesforce-einstein-mcp-server-by-cdata.git cd salesforce-einstein-mcp-server-by-cdata mvn clean install 2. Download, install, and license the CData JDBC Driver for Salesforce Einstein following the CData documentation. 3. Run the driver JAR to open the Connection String utility, configure your credentials, test the connection, and copy the connection string. 4. Create a .prp property file specifying Prefix, ServerName, ServerVersion, DriverPath, DriverClass, and JdbcUrl. 5. In claude_desktop_config.json, add the server entry: json { "mcpServers": { "salesforce-einstein": { "command": "PATH\\TO\\java.exe", "args": [ "-jar", "PATH\\TO\\CDataMCP-jar-with-dependencies.jar", "PATH\\TO\\salesforce-einstein.prp" ] } } } 6. Restart Claude Desktop completely to load the tools.

What you can do with Salesforce Einstein by CData

• Discover available Salesforce Einstein data tables by requesting schema overviews via conversational prompts. - Inspect specific column names and field types on customer account or opportunity objects. - Run read-only SQL SELECT queries to analyze correlations between won opportunities and account industries. - Retrieve open ticket counts and support project performance metrics using natural language questions. - Extract scheduled calendar events and upcoming customer appointments directly into local LLM sessions.

Key facts

• https://github.com/CDataSoftware/salesforce-einstein-mcp-server-by-cdata
• CRM, ERP & E-commerce, Databases & Data Stores
• crm, data-query, einstein, jdbc, salesforce

Part of MCP Servers

What is Salesforce Einstein by CData?

Salesforce Einstein by CData is a read-only Model Context Protocol server that bridges Large Language Models to live Salesforce Einstein systems. It operates using a CData JDBC driver to present Einstein entities as relational tables, allowing AI assistants to query enterprise data with standard SQL without permitting data modification.

Which MCP clients work with Salesforce Einstein by CData?

The server uses standard input and output for transport, which means it works with any MCP-compliant client that supports local stdio processes and runs on the same machine as the Java runtime. Claude Desktop is explicitly documented and supported for local integration.

What tools are provided by Salesforce Einstein by CData?

The server exposes three main tools: get_tables to retrieve available tables in CSV format, get_columns to inspect column definitions for a specified table, and run_query to execute custom SQL SELECT statements against the connected Salesforce Einstein instance.

Is Salesforce Einstein by CData open source?

Yes, the MCP server wrapper code is licensed under the open-source MIT License. However, connecting to the data source requires the CData JDBC Driver for Salesforce Einstein, which requires its own software license or trial from CData.