Amazon Athena

You can query live Amazon Athena data from LLMs through a local MCP server. This server wraps the CData JDBC Driver for Amazon Athena and exposes data via a simple MCP interface, enabling natural language questions about your data without writing SQL.

How to use

Once the MCP server is running, you interact with it through an MCP client. You don’t need to issue SQL queries directly. Ask questions like “What is the total revenue by region for last quarter?” or “Show open opportunities by account industry,” and the client will translate your request into MCP tool calls that run against your Athena data.

How to install

Prerequisites: Install Java Runtime Environment (JRE) and a Java Development Kit (JDK) sufficient to run the MCP server. You will also build the MCP server package with a build tool (Maven). After building, you will run the server with a local configuration file.

# 1) Clone the project
git clone https://github.com/cdatasoftware/amazon-athena-mcp-server-by-cdata.git
cd amazon-athena-mcp-server-by-cdata

# 2) Build the MCP server
mvn clean install

# This creates the runnable jar: CDataMCP-jar-with-dependencies.jar

# 3) Install the CData JDBC Driver for Amazon Athena from: https://www.cdata.com/drivers/athena/download/jdbc

# 4) License the JDBC driver (example paths shown for reference)
# Windows
# java -jar cdata.jdbc.amazonathena.jar --license
# Mac/Linux
# java -jar cdata.jdbc.amazonathena.jar --license

# 5) Configure a JDBC connection (writer-friendly steps)
# Run the connection string utility and copy the resulting JDBC URL into amazon-athena.prp later
java -jar cdata.jdbc.amazonathena.jar

# 6) Create a .prp file for your JDBC connection (example)
# amazon-athena.prp content (replace PATHs with your setup)
# Prefix=amazonathena
# ServerName=CDataAmazonAthena
# ServerVersion=1.0
# DriverPath=PATH\TO\cdata.jdbc.amazonathena.jar
# DriverClass=cdata.jdbc.amazonathena.AmazonAthenaDriver
# JdbcUrl=jdbc:amazonathena:InitiateOAuth=GETANDREFRESH;
# Tables=

# 7) Run the MCP server using the .prp file
java -jar CDataMCP-jar-with-dependencies.jar /PATH/TO/amazon-athena.prp

# 8) Optional: integrate with Claude Desktop by following the configuration steps for your environment

Additional sections

Configuration and usage notes: The MCP server runs locally (stdio) and can be used with clients on the same machine. You will typically create a configuration file that describes the JDBC connection, then start the MCP server with that file. The server exposes a small set of tools for listing available tables and columns and for executing read-only queries.

Usage patterns include:

• Asking for a list of tables in your Athena data source and then listing columns for a chosen table.
• Running a SELECT query that filters data, for example to fetch a subset of records for a report.
• Letting the MCP client handle routing of your natural language questions to the appropriate SQL behind the scenes.

Troubleshooting tips:

• Ensure the MCP server process is running and the console shows it listening on standard input/output.
• Verify the amazon-athena.prp file contains the correct JDBC URL and driver class.
• If data cannot be retrieved, confirm the JDBC driver is licensed and the connection string works by testing it with the driver utility before starting the MCP server.

Available tools

amazonathena_get_tables

Retrieves a list of tables available in the data source. Output is CSV with column headers.

amazonathena_get_columns

Retrieves a list of columns for a specific table. Output is CSV with column headers.

amazonathena_run_query

Executes a SQL SELECT query against the data source.