Redshift MCP Server (TypeScript)

What is Redshift MCP Server (TypeScript)?

It is a Model Context Protocol (MCP) server for Amazon Redshift, implemented in TypeScript. It enables LLMs in Cursor IDE and other MCP-compatible clients to inspect database schemas and execute read-only queries.

How to use Redshift MCP Server (TypeScript)?

Configure the server in a .cursor/mcp.json file (project‑specific or global) with a DATABASE_URL environment variable following the redshift://username:password@hostname:port/database?ssl=true format. Install dependencies with npm install, build with npm run build, and run via npm start or directly with node dist/index.js. For development use npm run dev.

Key features of Redshift MCP Server (TypeScript)

• Execute read‑only SQL queries with the query tool.
• Get detailed table structure with the describe_table tool.
• Find tables containing columns matching a name pattern using find_column.
• Automatically discover and list schemas, tables, and columns.
• Provide sample rows (up to 5) with automatic sensitive‑data redaction.
• Use read‑only transactions and input sanitisation for security.

Use cases of Redshift MCP Server (TypeScript)

• Inspect database schemas and table structures through natural language.
• Run ad‑hoc read‑only queries (e.g., count orders by status).
• Search for columns across all tables (e.g., find columns containing “email”).
• Retrieve sample data from tables without exposing sensitive information.
• Obtain table statistics like size, row count, and creation time.

FAQ from Redshift MCP Server (TypeScript)

What are the prerequisites?

Node.js 16 or higher, TypeScript, access to an Amazon Redshift cluster, and Cursor IDE installed.

How do I connect to my Redshift database?

Set the DATABASE_URL environment variable to a Redshift‑formatted connection string: redshift://username:password@hostname:port/database?ssl=true.

Is the server read‑only?

Yes. The server uses read‑only transactions for queries to prevent modifications and sanitises inputs to avoid SQL injection.

Where does the data live?

All data stays in your own Amazon Redshift cluster. The server connects directly to it and does not store or expose data outside your environment.

What transport and authentication does it use?

The server uses stdio transport. Authentication is handled via the Redshift connection URL (username/password). No additional MCP‑level authentication is required.