UBOS Asset Marketplace: CData MCP Server for SingleStore – Unleash the Power of Your Data with AI

In the rapidly evolving landscape of artificial intelligence and data-driven decision-making, the ability to seamlessly connect Large Language Models (LLMs) with real-time data is becoming increasingly critical. UBOS is at the forefront of this transformation, offering a comprehensive AI Agent Development Platform designed to empower businesses across all departments. Our platform simplifies the orchestration of AI Agents, facilitates connections to enterprise data, and enables the creation of custom AI Agents tailored to your specific needs.

Within the UBOS ecosystem, the Asset Marketplace plays a pivotal role in providing pre-built integrations and tools that accelerate AI adoption. One such asset is the CData MCP Server for SingleStore, a powerful solution that bridges the gap between SingleStore databases and LLMs like Claude Desktop.

Understanding the CData MCP Server for SingleStore

The CData MCP Server for SingleStore is a Model Context Protocol (MCP) server that allows you to connect to SingleStore data from Claude Desktop through CData JDBC Drivers. This integration enables you to query live SingleStore data using natural language, eliminating the need for complex SQL queries and empowering users with varying technical expertise to access valuable insights.

Key Features:

• Seamless Integration: Connects SingleStore data to Claude Desktop and other MCP-compatible LLMs.
• Natural Language Queries: Allows users to query data using natural language, simplifying data access.
• Real-Time Data Access: Provides access to live SingleStore data, ensuring that insights are based on the most up-to-date information.
• Simplified Setup: Offers a straightforward setup process, allowing users to quickly connect to their SingleStore data.
• Read-Only Access (This Version): This specific version provides read-only access to SingleStore data, ensuring data integrity.

Important Note: This project builds a read-only MCP server. For full read, write, update, delete, and action capabilities and a simplified setup, check out the free CData MCP Server for SingleStore (beta).

The Power of MCP: A Standardized Bridge for AI and Data

Before diving deeper, let’s understand the core technology behind this asset: MCP. The Model Context Protocol (MCP) is an open standard that streamlines how applications furnish context to LLMs. Essentially, an MCP server acts as an intermediary, enabling AI models to access and interact with external data sources and tools, like your SingleStore database.

Imagine MCP as a universal translator, allowing LLMs to understand and utilize data from various sources without requiring intricate custom integrations for each. This standardization accelerates AI development and deployment, making it easier to leverage the power of LLMs for diverse business applications.

Use Cases: Transforming Data into Actionable Insights

The CData MCP Server for SingleStore unlocks a wide range of use cases by enabling LLMs to access and interpret data stored in SingleStore databases. Here are a few examples:

• Data-Driven Decision Making: Empower executives and managers to make informed decisions by querying real-time SingleStore data using natural language. For example, a marketing manager could ask, “What is the correlation between closed won opportunities and the account industry?” to identify high-performing market segments.
• Customer Support Automation: Integrate SingleStore data with customer support AI Agents to provide personalized and efficient support. An AI Agent could access customer information, support tickets, and product details to quickly resolve customer issues.
• Sales Performance Analysis: Analyze sales data in SingleStore to identify trends, track performance, and optimize sales strategies. A sales manager could ask, “How many open tickets do I have in the SUPPORT project?” to assess the workload of the support team and identify areas for improvement.
• Calendar and Event Management: Connect SingleStore data with calendar and event management systems to provide AI Agents with access to scheduling information. An AI Agent could answer questions like, “Can you tell me what calendar events I have today?”
• Automated Report Generation: Generate automated reports based on SingleStore data, saving time and resources. An AI Agent could be instructed to create a weekly sales report summarizing key performance indicators.

Setting Up the CData MCP Server for SingleStore: A Step-by-Step Guide

Integrating the CData MCP Server for SingleStore into your workflow is a straightforward process. Here’s a detailed guide:

1. Clone the Repository:

   Begin by cloning the repository from GitHub:

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

2. Build the Server:

   Build the server using Maven:

   bash mvn clean install

   This command generates the CDataMCP-jar-with-dependencies.jar file.

3. Download and Install the CData JDBC Driver:

   Download and install the CData JDBC Driver for SingleStore from https://www.cdata.com/drivers/singlestore/download/jdbc.

4. License the CData JDBC Driver:

   • Navigate to the lib folder within the installation directory (e.g., C:Program FilesCDataCData JDBC Driver for SingleStore on Windows or /Applications/CData JDBC Driver for SingleStore/ on Mac/Linux).
   • Run the command java -jar cdata.jdbc.singlestore.jar --license.
   • Enter your name, email, and “TRIAL” (or your valid license key).

5. Configure Your Connection to SingleStore:

   • Execute the command java -jar cdata.jdbc.singlestore.jar to launch the Connection String utility.
   • Configure the connection string, ensuring you test the connection to verify its success. If your SingleStore data source uses OAuth, you’ll need to authenticate via your web browser.
   • Once the connection is successfully tested, copy the connection string for the next step.

6. Create a .prp File for Your JDBC Connection:

   Create a .prp file (e.g., singlestore.prp) with the following properties, replacing the placeholder values with your actual configuration:

   env Prefix=singlestore ServerName=CDataSingleStore ServerVersion=1.0 DriverPath=PATHTOcdata.jdbc.singlestore.jar DriverClass=cdata.jdbc.singlestore.SingleStoreDriver JdbcUrl=jdbc:singlestore:InitiateOAuth=GETANDREFRESH; # Replace with your actual JDBC connection string Tables=

   • Prefix: A prefix used for the exposed tools.
   • ServerName: A descriptive name for your server.
   • ServerVersion: The version number for your server.
   • DriverPath: The full path to the CData JDBC Driver JAR file.
   • DriverClass: The JDBC Driver Class name (e.g., cdata.jdbc.singlestore.SingleStoreDriver).
   • JdbcUrl: The JDBC connection string copied from the Connection String utility.
   • Tables: Leave blank to access all tables or specify a comma-separated list of tables to expose.

Integrating with Claude Desktop

To use the MCP Server with Claude Desktop, follow these steps:

1. Create or Modify the Claude Desktop Configuration File (claude_desktop_config.json):

   Add a new entry for the CData MCP Server to the mcpServers section of the configuration file. The file location varies by operating system:

   • Windows: %APPDATA%Claudeclaude_desktop_config.json
   • Linux/Mac: /Users/{user}/Library/Application Support/Claude/claude_desktop_config.json

   Here’s an example configuration snippet:

   Windows:

   { “mcpServers”: { “singlestore_mcp”: { “command”: “PATHTOjava.exe”, “args”: [ “-jar”, “PATHTOCDataMCP-jar-with-dependencies.jar”, “PATHTOsinglestore.prp” ] } } }

   Linux/Mac:

   { “mcpServers”: { “singlestore_mcp”: { “command”: “/PATH/TO/java”, “args”: [ “-jar”, “/PATH/TO/CDataMCP-jar-with-dependencies.jar”, “/PATH/TO/singlestore.prp” ] } } }

   Replace PATHTOjava.exe (Windows) or /PATH/TO/java (Linux/Mac) with the actual path to your Java executable. Similarly, update the paths to CDataMCP-jar-with-dependencies.jar and singlestore.prp.

   Important: Ensure that you use a unique classname_dash value for each MCP server configured in the mcpServers section. Replace {classname_dash} with a suitable identifier, like singlestore_mcp.

2. Restart Claude Desktop:

   Fully exit and restart Claude Desktop for the changes to take effect. You may need to use Task Manager (Windows) or Activity Monitor (Mac) to ensure the application is completely closed.

Running the MCP Server Independently

For testing or debugging purposes, you can run the MCP Server directly from the command line:

bash java -jar /PATH/TO/CDataMCP-jar-with-dependencies.jar /PATH/TO/singlestore.prp

Note: This server uses stdio and is intended for clients running on the same machine.

Interacting with SingleStore Data Through AI

Once configured, your AI client (like Claude Desktop) can leverage the built-in tools to interact with your SingleStore data. You generally don’t need to call these tools directly; simply ask the client natural language questions. Here are a few examples:

• “What is the correlation between my closed won opportunities and the account industry?”
• “How many open tickets do I have in the SUPPORT project?”
• “Can you tell me what calendar events I have today?”

Available Tools

The MCP Server exposes the following tools (where {servername} corresponds to the name defined in your configuration file, e.g., singlestore_mcp):

• {servername}_get_tables: Retrieves a list of tables available in the SingleStore database. Use {servername}_get_columns to list columns for a specific table. Output is in CSV format.
• {servername}_get_columns: Retrieves a list of columns for a given table. Use {servername}_get_tables to find available tables. Output is in CSV format.
• {servername}_run_query: Executes a SQL SELECT query against the SingleStore database.

Troubleshooting Common Issues

• Server Not Visible in Claude Desktop: Ensure you’ve completely quit and restarted Claude Desktop.
• Data Retrieval Issues: Double-check your connection string in the .prp file. Use the Connection String builder to verify its accuracy.
• Connection Problems: Contact the CData Support Team for assistance with data source connectivity.
• MCP Server Issues or Feedback: Join the CData Community for support and to share your thoughts.

License Information

This MCP server is released under the MIT License, offering you the freedom to use, modify, and distribute the software according to the license terms. Refer to the LICENSE file within the project repository for comprehensive details.

UBOS: Your Partner in AI-Powered Transformation

The CData MCP Server for SingleStore is just one example of how UBOS empowers businesses to leverage the power of AI. Our full-stack AI Agent Development Platform provides the tools and infrastructure you need to build, deploy, and manage AI Agents across your organization. With UBOS, you can:

• Orchestrate AI Agents: Seamlessly manage and coordinate multiple AI Agents to automate complex tasks and workflows.
• Connect to Enterprise Data: Integrate AI Agents with your existing data sources, including databases, CRM systems, and cloud applications.
• Build Custom AI Agents: Develop custom AI Agents tailored to your specific business needs, using your own LLM models and data.
• Create Multi-Agent Systems: Design and deploy sophisticated AI systems that leverage the collective intelligence of multiple AI Agents.

By leveraging the UBOS platform and assets like the CData MCP Server for SingleStore, you can unlock the full potential of your data and transform your business with AI.