amazon-redshift-mcp-server-by-cdata

What is amazon-redshift-mcp-server-by-cdata?

Amazon-redshift-mcp-server-by-cdata is a read-only MCP (Model Context Protocol) server built by CData that allows LLMs like Claude Desktop to query live data from Amazon Redshift using natural language. It wraps the CData JDBC Driver for Amazon Redshift and exposes the data as relational SQL models through a simple MCP interface, requiring no manual SQL from the user.

How to use amazon-redshift-mcp-server-by-cdata?

Clone the repository, build with mvn clean install, download and install the CData JDBC Driver for Amazon Redshift, license it, create a connection string using the driver’s utility, then configure a .prp file with connection properties. Finally, add the server to your Claude Desktop config (claude_desktop_config.json) using the built JAR and the .prp file, then restart Claude Desktop.

Key features of amazon-redshift-mcp-server-by-cdata

• Read-only access to Amazon Redshift data via MCP
• Works with LLMs like Claude Desktop for natural language queries
• Exposes tools: {server}_get_tables, {server}_get_columns, {server}_run_query
• Uses CData JDBC Driver for connectivity
• Supports all tables or explicitly declared tables
• Communicates via stdio transport (same machine only)

Use cases of amazon-redshift-mcp-server-by-cdata

• Letting a business analyst ask “What is the correlation between closed won opportunities and account industry?” without writing SQL
• Enabling an AI assistant to answer “How many open tickets do I have in the SUPPORT project?” from Redshift data
• Providing real-time Redshift querying to LLM-based chat interfaces for ad‑hoc data exploration

FAQ from amazon-redshift-mcp-server-by-cdata

What is the difference between this read-only MCP server and the full CData MCP Server for Amazon Redshift?

This server is read-only; it only supports SQL SELECT queries. For full read, write, update, delete, and action capabilities, CData offers a separate free beta MCP Server for Amazon Redshift with a simplified setup.

What are the runtime requirements?

You need Java (to run the JAR), Maven (to build from source), and the CData JDBC Driver for Amazon Redshift (downloaded and licensed separately). The driver must be installed and its JAR path provided in the .prp file.

Where does the data live?

Data stays in your Amazon Redshift instance. The MCP server connects to it via the JDBC driver using a connection string; no data is copied or stored locally by the server itself.

What transport does the server use?

The server uses stdio as its transport layer, meaning it can only be used with clients running on the same machine as the server (e.g., Claude Desktop locally).

How is authentication handled?

Authentication is configured in the JDBC connection string (e.g., InitiateOAuth=GETANDREFRESH for OAuth). You use the driver’s Connection String utility to test and copy the final connection string into the .prp file.