home / mcp / redis mcp server

Redis MCP Server

Provides a read-only MCP interface to query live Redis data via the CData JDBC Driver.

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "cdatasoftware-redis-mcp-server-by-cdata": {
      "command": "java",
      "args": [
        "-jar",
        "/PATH/TO/CDataMCP-jar-with-dependencies.jar",
        "/PATH/TO/redis.prp"
      ]
    }
  }
}

You can run a local, read-only MCP server that exposes Redis data through a simple MCP interface. This lets an AI client query live Redis data without writing SQL, enabling natural language questions about your data while keeping the data access read-only on your machine.

How to use

Once the MCP server is running, you connect your MCP client or Claude Desktop-style tool to the local server. The server exposes tools that let you list available data, inspect table schemas, and run read-only queries using familiar SQL syntax under the hood. You can ask questions like “What is the total count of records in Account where IsDeleted is true?” or “Show me the columns available in the Account table.” The client will use the built-in tools to discover tables and columns, then execute read-only queries to retrieve live Redis data through the CData JDBC Driver for Redis.

You typically won’t need to call the tools directly. Instead, you ask natural language questions and the AI client will leverage the exposed MCP tools to fetch the data needed to answer your question.

How to install

Prerequisites: Java Runtime Environment (JRE) installed on your machine. Ensure you have the JRE available as java in your system PATH.

1. Build the MCP server package from its source code.

git clone https://github.com/cdatasoftware/redis-mcp-server-by-cdata.git
cd redis-mcp-server-by-cdata
mvn clean install

2. Prepare the JDBC driver and the configuration file for a Redis connection. You will install the CData JDBC Driver for Redis separately, license it, and create a .prp file that defines the JDBC connection to Redis. Follow the steps to download, license, and configure the driver, then create a Redis.prp file with your connection details.

3. Run the MCP server using the prepared .prp file.

4. Start the server with the final run command.

Additional notes and configuration

The server runs in stdio mode, meaning it is intended to be used with an MCP client on the same machine. The basic run command looks like this: java -jar CDataMCP-jar-with-dependencies.jar /PATH/TO/redis.prp. This launches the MCP server using the Redis data source described in the redis.prp file.

You can configure Claude Desktop (or another MCP client) to start the MCP server by pointing to the same java command and passing the path to your redis.prp file in the arguments. If you need to refresh the client after changes, fully quit and re-open the client so the MCP server becomes available again.

Troubleshooting

If you cannot see the MCP server in your client, ensure the client process is fully closed and re-run the client. If data cannot be retrieved, verify that the connection string in redis.prp is correct and that the CData JDBC Driver for Redis is properly licensed and accessible. If problems persist, contact CData Support or the CData Community for assistance.

Available tools

redis_get_tables

Retrieves a list of tables accessible in the data source. The output is in CSV format with column headers.

redis_get_columns

Retrieves a list of columns for a specified table. The output is in CSV format with column headers.

redis_run_query

Executes a SQL SELECT query against the configured data source.