amazon-redshift -by-cdata
This read-only MCP Server allows you to connect to Amazon Redshift data from Claude Desktop through CData JDBC Drivers. For full CRUD support, check out the first managed MCP platform: CData Connect AI (https://www.cdata.com/ai/).
claude mcp add --transport stdio cdatasoftware-amazon-redshift-mcp-server-by-cdata java -jar /PATH/TO/CDataMCP-jar-with-dependencies.jar /PATH/TO/amazon-redshift.prp \ --env LD_LIBRARY_PATH="Optional: set if JDBC driver requires native libs" \ --env JAVA_TOOL_OPTIONS="Optional: set if needed"
How to use
This MCP server exposes Amazon Redshift data via a read-only MCP interface built on top of the CData JDBC Driver for Amazon Redshift. It wraps the JDBC driver so language models can query live Redshift data using natural language questions without writing SQL. Clients, such as Claude Desktop or other MCP-aware tools, connect to the server using the configured MCP entry (defined in the claude_desktop_config.json-like structure) and can invoke tools to list tables, describe columns, or run queries. The server operates in stdio mode, meaning it is intended for local clients on the same machine as the server and supports a set of tools named after the server (e.g., amazon_redshift_get_tables, amazon_redshift_get_columns, amazon_redshift_run_query) which return results in CSV or structured formats suitable for downstream processing. Example queries include retrieving available tables, inspecting columns of a table, or executing a SELECT statement against the data source.
How to install
Prerequisites:
- Java Runtime/SDK (JDK 8+ recommended)
- Maven (for building the MCP server)
- Access to a CData JDBC Driver for Amazon Redshift and a valid license
Installation steps:
-
Clone the repository and navigate into it: git clone https://github.com/cdatasoftware/amazon-redshift-mcp-server-by-cdata.git cd amazon-redshift-mcp-server-by-cdata
-
Build the MCP server JAR: mvn clean install This creates the JAR file named CDataMCP-jar-with-dependencies.jar (path as output shows in your build).
-
Install and license the CData JDBC Driver for Amazon Redshift:
- Download the driver from https://www.cdata.com/drivers/redshift/download/jdbc
- Install according to the platform guidelines (Windows/macOS/Linux)
- License the driver, typically by running a command like: java -jar cdata.jdbc.redshift.jar --license (Follow prompts to enter name, email, and TRIAL or license key)
-
Create a .prp file for your JDBC connection (e.g., amazon-redshift.prp) with the required properties: Prefix=redshift ServerName=CDataRedshift ServerVersion=1.0 DriverPath=PATH\TO\cdata.jdbc.redshift.jar DriverClass=cdata.jdbc.redshift.RedshiftDriver JdbcUrl=jdbc:redshift:InitiateOAuth=GETANDREFRESH; Tables=
-
Run the MCP server: java -jar /PATH/TO/CDataMCP-jar-with-dependencies.jar /PATH/TO/amazon-redshift.prp
Note: The server uses stdio and is intended for local clients on the same machine.
-
Configure your client (e.g., Claude Desktop) with the appropriate mcpServers entry, pointing to the java executable, the JAR, and the path to the .prp file as described in the README example.
Additional notes
Tips and troubleshooting:
- Ensure the CData JDBC Driver is properly licensed before starting the MCP server.
- The MCP server runs in stdio mode; use local clients or ensure your client supports stdio-based MCP servers.
- If Claude Desktop cannot see the server, fully quit and relaunch the client to refresh the MCP server list.
- Double-check the connection string (JdbcUrl) in your .prp file to ensure it points to the correct Redshift cluster and that any OAuth or authentication steps are satisfied.
- The server name used in client config typically follows the {servername}_get_tables, _get_columns, and _run_query tool naming convention; keep the server key consistent (e.g., amazon_redshift) across the config and tool names.
- When moving between environments (dev/stage/prod), update the .prp JdbcUrl and any DriverPath references accordingly.
